> * I want to be able to simulate GR.
   > * I want to understand QFT
   > * I want to be able to simulate atoms and molecules from first-principles.

With the current technology, assuming standard model of particle physics (which is QFT) is the first-principle, simulating atoms would take trillions of the age of the universe, let alone molecules, or any effect of GR.

If you haven't already, I highly recommend Feynman's Lectures on Physics: http://feynmanlectures.caltech.edu

While much more advanced than typical introductory texts, it does provide deeper insights than typical introductory texts, which makes it worth the investment.

I'm a beginner, thinking about returning to school for physics, but I'm dipping in with this tome I purchased recently. He's a great explainer for someone at my level.

And I have to brush up on all of the math (and learn calculus...) in order to fully understand. It's exciting that reading him isn't just a cold hard wall. It's proving to be a great guide into what to sort of triage my studies.

I strongly second this one! (Though I have to admit that I purchased a not-cheap edition of the first volume to get started -- though I prefer reading from paper, and don't yet own an eReader and have a bit of a burgeoning library anyway. But thanks for the link! it's nice to have the rest on hand).

I'm more at the level of Goldstein for Mechanics, and learning about the Lagrangian. Of course, there's still much to learn from Feynman.