When the colors are true they tend to pop. Also, one notices subtle variations in color, say on a leg, a face, or a fold of cloth. These areas become monochromatic in copies where the colors have dulled.
It has been suggested that the strength and luminosity of the colors depends on whether a given copy was early or late in its print run, rather than on lack of color-fastness or sun fading. I tend to agree with this myself, and it seems to be the general consensus here.
But I have noticed that even when reds are intense, it does not necessarily follow that blues will be also.
I think it is feasible that blues can fade to a slightly greenish hue, just as they frequently do on Superman's costume, but that not all colors fade equally.
There is also the possibility that color overlays are intentionally used to modify primary colors.
For example, pure yellow backgrounds may sometimes have had a reddish overlay. So it doesnt always follow that the brightest colors are "correct" in the sense that they follow the intended color composition.
I think you are also probably quite right to point out that it doesnt always follow that if an area of color is now orange, that it was always intended to be red. And in some cases we may never know.
When one does see a cover as it was intended, though, there is generally no mistaking it is there? The vast majority of Fiction House are not scarce relative to many other Golden Age books (except in a small handful of cases). But combine low print runs with "original" colors and scarce becomes rare, even fugitive.