If Pathfinder 2 fails it's probably because of 5E or their class design.
I would want to discuss this in more detail, since "class design" could mean pretty much anything.
If Pathfinder 2 fails, I would point to some or all of the following factors:
1) Pathfinder 2 seemingly ignores that 5E succeeds because it abandons fiddliness
1a) two important factors where the jury is still out are caster-martial balance and ease of monster prep for the GM
1b) Paizo might completely underestimate how much people actually like bounded accuracy
2) Pathfinder 2 seemingly ignores the deep impopularity of the 4E-style design of lots of small options (yes, you could phrase this as "all options are balanced" or "no trap options", but players seem to like the existence of some options being better than others since that makes their decisions matter)
Still want that overhauled and fixed 3.5 type game.
I would argue 5E
is the fix to 3E.
I believe there no longer exist a substantial market for a more direct descendant to 3E. The core engine of d20 simply is so broken it can't be fixed, or rather, that 5E proved it could be fixed, only that you had to really reassemble the fundamentals to do it.
Yes, 5E also adds a number of simplifications that has nothing to do with fixing 3E. Some of them are unnecessary or even outright dumb.
I remain impressed over just how thoroughly 5E solves many of the deal-breaking deficiencies that forced me to leave 3E. I honestly don't believe a more direct "create a 3.999 edition" would have had the courage and energy to truly uproot the core problems with the d20 engine, instead creating just another iteration in the 3.0 - 3.5 - PF chain that never came close to any of 5E's true fixes, so I think "that overhauled and fixed 3.5 type game" is mostly a pipe dream that could never have happened.