Great post, Erick. One nit:
diffraction, the bending of light around small openings.
That is only true when the leading order (Fraunhofer) approximation vanishes, such as in the rarely supernumeraries in rainbows, not in typical circumstances such as optics, photography, etc., so in our case: Diffraction is a range of field intensities from the superimposition of propagating waves. In fact, diffraction occurs even without an aperture, such as when of multiple radio antennas (even without a parabolic dish) are spread out over a large area, and used for radar or imaging distant galaxies, etc. In that case, there are no edges to bend around, but the diffraction still occurs the same.
Another way to think of it is that diffraction is as an effect of the Heisenberg Uncertainty Principle. The principle says that there is a limit to the precision with which you can simultaneously know both the position and the momentum of a particle. When you reduce the size of the lens aperture, you are increasing the knowledge of the position of the photons: you know even more accurately what space they went through, because the opening is smaller. But the price you pay is you now know less about their direction: they go in more random directions.