Don't worry about "support heavy." Do worry about Runepriest. And read up on the pedagogical technique of scaffolding. And go easy on the minions (even accounting for encounter for XP budgets).
On your character mix: fighter and warlords are great. Runepriests are contraindicated.
The non-hybrid fighter1 and warlord will be best buds, so long as they avoid the obvious character building "traps."2 They work very well together and the classes themselves have plenty of interesting and powerful options.
The runepriest is delightful in concept, and horrible in execution. It really should come with a label "for experts only" as it's a great way to completely distract an expert optimiser by getting their performance back on par with a "normal" group. To quote the runepriest handbook:
The Runepriest is one of the least well-supported classes in 4e, with only its initial book and a single Dragon article of note giving it anything to work with. Despite that, it's strong enough on its own merits to actually be worth something . . . but what? ... All Runepriest at-will and encounter attacks are actually two powers in one. With each attack, you get to choose whether to apply the rune state of destruction or the rune state of protection. Not only will this change your rune state (and thus the passive bonuses you're giving your allies with your mere presence), but it'll also change what the power itself does, often pretty drastically. While most powers are going to be more useful for one rune state or the other, plenty of powers are still great for both options, giving you some turn-by-turn flexibility that most 4e classes can't match.
The runepriest is a stateful class, requiring huge amounts of bookkeeping relative to other classes, and requires huge amounts of battlefield awareness due to the choices it must make every turn. It's possible to make a adequate one, but it's not something that can be produced without extensive consideration and system mastery.
On the other hand, the idea of double-leader (warlord and shaman) + fighter is not a problem at all, especially if the leaders are granting attacks to the fighter. (It's in fact, one of the better party builds from a theoretical standpoint.)
Don't worry about providing NPCs, just scale back the number of enemies in an encounter a little bit.
Provide for needed skills and good power synergies by offering pre-chosen good "menus" of builds.
In general, the best help I've found is by gently encouraging people choosing powers away from the obviously poor options. Look at the handbook index and provide a smaller menu of choices for your players when they build their characters. Beyond that, run the characters through the DDI character builder to make sure the maths come out well.
Environmental and dungeon traps and hazards do not require dedicated character classes to deal with and, in the worst case, can be bashed with a stick until they stop attacking. Encourage a wide skill selection (and multiclassing) such that every skill is nominally represented (and avoid creating skill challenges for skills that aren't taken by the party.)
There will be no problems with non-combat encounters (skill challenges), as the choices the party makes in overcoming them will reflect the choices the party made during character creation. I would recommend reading much of the literature available on skill challenges as they can be very easy to run poorly.
Crowd control at low-heroic is certainly useful, but not required. The party's natural focus will be on individual monsters rather than clearing mobs of minions. Don't worry too much about it, as some leaders can make fantastic off-controllers if they choose to embrace that path. Furthermore double-leader + defender should be able to, in any given encounter, find a nice defensible spot to hole up and deal with the crowd one at a time.
Don't mistake 4e for 3.5. Spend some time looking at pedagogical questions on the stack.
As a new 4e DM coming from 3e, take a look at the guide to convert a 3.5 DM to a 4e DM. I will reemphaise my point to not simulate, as that's not a design goal and is not supported.
To teach your players take a look at:
I would absolutely allow arbitrary character "retcons" until everyone is happy with their character and role in the party. By making no cost to a retcon (and by having magic items just be a part of character, according to the standard uncommon of level+1, uncommon of equal level, uncommon of level-1, and cash equal to the cost of a common of level-1), your players can explore the mechanics of the game to a degree that doesn't impact the story, thereby allowing a real story to develop without the "accidental" character deaths of people dissatisfied that their expectations of character doesn't match the performance.Scaffolding is important,
Scaffolding your players' learning is important.
In order to effectively teach your new players, make sure that each new element introduced is just outside their mastered skills zone. As you introduce concepts one by one, make sure to always be challenging the characters with a singular new concept. By providing scaffolding (as Van Der Stuyf 2002 puts it):
“The zone of proximal development is the distance between what children can do by themselves and the next learning that they can be helped to achieve with competent assistance” (Raymond, 2000, p.176). The scaffolding teaching strategy provides individualized support based on the learner’s ZPD (Chang, Sung, & Chen, 2002). In scaffolding instruction a more knowledgeable other provides scaffolds or supports to facilitate the learner’s development. The scaffolds facilitate a student’s ability to build on prior knowledge and internalize new information.
Therefore, much of the cognitive work of "what agency does my character have?" will start with menus you provide your players. By starting highly scaffolded and gradually showing why everything the way it is, your players can be taught system mastery far more effectively than having the rules compendium and first essentials book shoved in their faces.
A sample "menu" for your players:
The job of a DM, in this instance, is to act as a sommelier: "Why yes, Ma'am, that feat will complement your chosen power wonderfully. If I may suggest, Sir, this item instead?..." (Bow tie optional.) The actual menus I was working on were getting a little long for this answer, so I've moved them to a google doc. The trick is that your players should use the mechanical chassis presented in the menus as inspiration for whatever fluff they feel fits their desired character. And to build in narrative party cohesion as well as mechanical synergies.
In my initial five or so readings of your question, my mind completely skipped over your mention of a "hybrid" fighter-striker, as some sort of wrong-jargony attempt to describe a fighter dedicated to doing damage. If you're actually
considering a hybrid, don't. There are plenty of ways of giving a fighter highly adequate damage without hybridization. If you absolutely must hybridize, look at the various handbooks
to help you avoid the obvious build problems, the subtle build problems, and the build problems that only appear in play. Speaking as someone who has tried to play a hybrid controller-leader in mid-epic play, just don't do it.
2 There are many powers in D&D 4th edition. Some are better than others. Some are phrased or written so poorly that they are out-performed by at-wills. These are traps, in the Admiral Ackbar sense.