No, Pathfinder actually makes the balance quite a bit worse.
The weakest classes got weaker, and the strongest classes got stronger.
Changes to combat feats and combat maneuvers mean that the few mundane tricks the system had, are now much more difficult to pull off and are less useful when you do.
Changes to magic items disproportionately hurt martial characters
Insistence on strict and painful “realism taxes” on mundane characters, even well into high levels, prevent them from keeping up.
Nerfs to spells were few and haphazard – while a few specific cases were nerfed considerably, most were untouched, and plenty of high-power options remain
Class features for spellcasters are occasionally quite potent. While some archetypes for mundane classes are also pretty good, the fact that the best classes could actually get substantial improvements is worrisome.
Combat Maneuvers
The math on Combat Maneuver Bonus vs. Combat Maneuver Defense is poor. Simplifying and unifying the systems was a good idea, but they failed on execution. Because you add both Str and Dex to CMD, but only one to CMB (and for several other reasons, this is just the simplest one), CMD scales much better than CMB does. As a result, at mid-to-high levels it becomes very difficult to actually use combat maneuvers.
Meanwhile, most of the Improved (combat maneuver) feats have been split into both Improved ____ and Greater ____: two feats to do what you used to be able to do in a single feat. In the most important case, Trip, two feats to not even do what you did with one in 3.5: now the follow-up attack off of a trip is an attack of opportunity, which means trippers require Combat Reflexes to use it, and need a lot more Dexterity.
Items
Changes to items are also serious problems to martial classes. All enhancement bonuses to a physical ability score are found only on belts – which means if you want more than one, you have to pay harsh premiums on combining items. Martial characters need bonuses to both Strength and Constitution (or Dexterity and Constitution in some cases), because the math of the system still has those bonuses baked into the design assumptions, but now they cost more. Compare this to 3.5, which, as of Magic Item Compendium, charged no premium to add these sorts of abilities even to other items entirely.
Meanwhile, while it’s also true that all enhancement bonuses to mental scores are headbands, most spellcasters only care about one mental score (Wis for clerics and druids, Cha for bards and sorcerers, Int for wizards, and so on), and then they are free to get a bonus to Constitution (most character’s second-most-important ability) without having to pay extra.
To add insult to injury, the official guidance on houseruling magic items explicitly says you should not allow these items in different slots even as a houserule! They were so adamant about enforcing this massive penalty on mundane characters that they literally called out DMs who were considering fixing it!
Realism Taxes
The most glaring example here, though far from the only one, is the insistence that, “just because it costs a feat,” i.e. a major character resource, doesn’t mean exotic weapons are supposed to be better than martial weapons. And, in fact, none of them are actually good. 3.5 had a lot of problems with Exotic Weapon Proficiency; very few of the exotic weapons were actually worth a feat. Pathfinder made very sure to nerf those so they no longer were.
Minimal spell nerfs
Depowering spellcasters really requires going through each and every spell; Paizo did not even attempt this. They did nerf a few particularly-notorious spells, but missed many others. They did nerf the polymorph spells, which desperately needed it, so that is a good thing; Pathfinder polymorphing is without a doubt a massive improvement over 3.5. But it’s also just about the best change in the entire system.
Further, they have printed new spells of absurd power. Do a google search for paragon surge and the ridiculous things that can be done with it.
Astonishing class features for spellcasters
The arcanist, a new Pathfinder class, is by-far the worst offender here, getting the best of both worlds between prepared and spontaneous spellcasting, off the best spell list in the game, plus free metamagic effects (read: one of the most powerful classes of effects in the game). But the core classes also occasionally got major upgrades as well.
By contrast, the default fighter gets a variety of small numerical bonuses. The fighter’s problem was never numbers – if a fighter can make his numbers count, you’re dead, 3.5 or Pathfinder. The problem was always that the fighter is remarkably poor at forcing his numbers into the game, if the competition didn’t actively cooperate with him. None of that changed.
This list is not exhaustive
This isn’t everything that Pathfinder did that exacerbated the mundane-vs-magic divide that was already massive in 3.5. It was, in fact, a very consistent effort across Pathfinder’s entire history. The Pathfinder FAQ and errata are chock full of the developers stripping away anything powerful or useful they accidentally gave to mundane characters. Pathfinder is therefore one of the most magically-dominated systems out there.