Mention those textbooks and a bit about why that subject may be useful to learn.
Operating Systems: Three Easy Pieces + Linux Device Drivers 3rd: All the cool tech in my opinion is made with systems-programming. Game-engines, Risk/Pricing Engines, Radar/GPS, Compilers, VM's, OS's, probably also Apples AR goggles are made with systems languages, so likely/sadly C and C++. These two books are real and practical. The Linux one is outdated, but it worked for me.
Oppenheimers books on Signals and Systems: Because signal-processing. And the Fourier-Transformation is the most unintuitive AND useful piece of mathematics I have used/seen. (Stochastic Integrals is a close second)
Differential Equations. ODE's, SDE's, PDE's. I don't really have a good book here sadly, but Differential Equations describe the world.
1st one being "Thinking fast and slow" by Daniel Kahneman - completely reshaped my model of how much we are in control of our thought process.
2nd one was "Thus spoke Zarathustra" by Nietzsche - which is a very difficult read. Overall Nietzsche's approach to philosophy just immediately resonated with me and sent me down a path of thought that's still the foundation of my world model today. The whole approach of seeing life as inherently meaningless so meaning can be chosen at will was a great thing to have growing up. You definitely don't need to read the book to get there though - just reading a summary of what Nietzsche is about will probably be enough.
Basically every humanities scholar complains about the complexity of the issue, like social of political, just because of them not knowing linear algebra as a tool. You can solve almost all problem using matrix and tensor.
Still waiting for the Physics one, though!
Not directly a textbook per se, but was a deeply educational book for me about designing digital spaces and thinking about interface metaphors from a real-world, architecture-driven perspective.