Global illumination is the next big thing in graphics, it's basically an algorithm that represents how light interacts with objects, creating shadows, caustics and other cool lightning phenomena, it's known how to do it however it's estimated it could take up to 100x the processing power to truly show it in all it's glory giving animated movie like photorealistic graphics. Games can be partially GI'ed however. We are in a sense at a plateau. I will try and dig out a Tim Sweeney video for you, he addresses pretty much what you are asking.
When we reach 8k screens @ 120Hz the human eye cannot distinguish any more depth than that, we have surround sound, and also 3d sound is breaking through. AI has some work to do, but yes a high end PC can handle pretty much all the current data that is thrown at it. The next stage is much more photorealism more sophisticated AI, and even more ambitious 3d worlds.
These worlds are not easy to create, many man hours of creating models and animating them are involved, every detail has to be emulated.
Before that textures, physics, AI, rigging, animation, scripting, sound were all dumbed down, then GPU's became more advanced in their architecture, system and GPU memory & GPUCPU and CPU power increased, API's improved(DirectX and OpenGL), operating systems improved.
.........
EDIT: Here the video,
https://www.youtube.com/watch?v=XiQweemn2_A
EDIT AGAIN: Here's another good video,
https://www.youtube.com/watch?v=00gAbgBu8R4