I think that the difficulty for Intel is that they don't seem to attract the best talent anymore like they did 20 years ago. A big part of the job of running a leading edge fab is solving a lot of very tricky technical problems, and a lot less of those people work at Intel than used to. There are feedback loops in these things. The top people want to work with other top people. To some degree you can try to find those people and just offer them tons of money, which is what seemingly happened with Jim Keller (on the logic rather than the fab side, but still), but then he went and quit shortly thereafter.

Without the right people at the company, you can dump almost infinite money at a problem and still not solve it.