Now let's talk about reality itself, and the consciousness you think you are using to observe it

It was definitely strange to learn that mixing all colors of light together makes white light while mixing all paint colors together makes black paint.

In fact, yellow never exists during the entire process.

I feel like you could argue yellow does exist in HSV. I'm not a professional in programming, working with light, or using HSV so I may be wrong but my amateur understanding of HSV is that the hue strongly correlates with the wavelength of the light omitted (with the obvious exception of looping red back around into blue through wavelengths that can't exist), the saturation strongly correlates with the purity of that wavelength compared to every other (uniform distribution across the spectrum means white, then only that one wavelength is the pure expression of that color), and value is the amplitude of the resulting wavelength(s).

In this framework you could set hue to be just yellow, make the saturation very pure (high saturation), then have whatever amplitude you need (whatever value). Of course most of the time in modern computing HSV values are going to be derived from and converted back to RGB values, but for a brief period of time between computations you could argue that the computer has an actual representation of "yellow" itself and not just its Red and Green components.

It’s exactly analogous to an audio three-band graphic equalizer: if you were deaf and could only detect sound by looking at the equalizer display, you couldn’t distinguish between someone playing through the full scale and someone playing a single three-note chord while adjusting the volumes of the low, mid, and high notes.

Your brain made it up.

Well, yeah, but our brains make up every color. Our experiences of color are nothing more than a facet of our brains' interpretations of light information. It's all in our heads.

Magenta is a primary color (not red). You can't mix non-magenta red paint with blue paint to get purple.

On the topic of messing around in Paint, there's something really cool you can do assuming you have a display with normal pixels.
Make a new paint document and color the left half perfectly red and the right half perfectly blue (#FF0000 and #0000FF). Make sure the colors are touching in the middle. If you look really close at the place the colors touch, there will be a tiny little black gap. If you do the opposite, with red on the right and blue on the left, the gap will not appear.

This is, of course, because of the physical layout of the pixels with R on the left and B on the right. By putting red on the left and blue on the right, we make the biggest possible subpixel dark zone.

That was just your poorly calibrated display.

I've wondered the same about dogs' sensitive sense of smell. What's that like?

Colors don’t actually exist physically. They are a creation of the mind.

There's a great old Vsauce video that goes over something like this but even crazier to think about: "is your red the same as my red?". How can you know if your red is the same color as what someone else calls red? We could both be looking at different colors, but we both call it red. Even crazier, how do you explain colors to someone born blind?

How would you prove to a cat that those colors exist?

A spectrum test.

Show a human red fading to infrared, or purple fading to ultraviolet, next to cameras that can detect them in false color. Those are colors we can't see, yet you can see they're there.

Theoretically, it'd be the same for a cat or dog.

Maybe that's why animals don't seem very interested in TV. They don't have the same cones as us

There's no strictly defined set of wavelengths for yellow. But you could pick a random wavelength and as long as most people say it's yellow, that's a yellow wavelength.