@x-calibar - Ah I see, thanks for the info! I might try this later, I have a couple of ideas that may end up looking nice

@husmus - There are easy to use commercial applications like Gigapixel AI (which has a trial), and open source apps that can use pre-trained models/deep learning to do the same with varying results. Real-ESRGAN has a free command prompt exe that has some pre-trained models for anime.
This looks like a basic web service using Real-ESRGAN: https://arc.tencent.com/en/ai-demos/imgRestore
Didn't test, but should probably still work. Searching for Real-ESRGAN vulkan should find the github link I used, and Topaz Labs has Gigapixel AI.

As for other apps and Photoshop... I took the processed images, and dropped them into various layers to play around with. Maybe ran more processes or manually edited some. Don't remember what I did exactly! But, usually entails choosing what I want out of the various layers, and hiding the rest.

This is interesting, could you tell us more about how this works?

still more partial to the original, but i do love how even the res enhanced version still looks like it stays consistent with the art style