Yes, I know that. And as I said, unless I'm mistaken, you're only offered the X and Y values to input. The image below illustrates. That still doesn't deal with the issue of where these values are to be obtained for a particular filter.
I'm not even certain how, or convinced that, doing this would make a difference. I'm not seeing how this would change the effective 'colour' of the red, green and blue filters (channels) that are used to recombine the separations. Within different colour spaces you have differing RGB primaries. But in all colour spaces, you still have full saturation 255 for each of the channels. That's what needs to be changed if the filters are not exactly full red, green and blue. I don't see how changing the primaries; effectively changing the boundaries of the space, alters the full saturation value of 255. I understand that Red at 255 in sRGB will be different from Red at 255 in ProPhoto but it's still 255, full red. And it's not necessarily the saturation that needs to be changed but perhaps the hue, as was pointed out earlier and why, I think it was Kirk, suggested using colour fill layers so that the hue of the filter could be changed if and as needed. If just the saturation level of the filter (channel) needs to be changed it would seem that would be more easily done via Kirk's method than by creating a new colour space.