Jethro,
Do not worry about processing - this discussion has no relevance to what you are doing...
It is about some definitions that need to be clarified.
The point I am making is the amount of light, collected by lens aperture that reaches the individual pixel is the same regardless of sensor size (of course, assuming the pixel size is the same), and it depends only on lens diameter and FL (F-number).
My cameras (60D, 450D) also have APS sensor and I am using lenses designed for FF (24x36mm sensor size).
Lens performance will always be compromised at the corners of the sensor (film, CCD...), and it will always be better in the centre.
So, smaller sensorr actually means you will not see those imperfections that would appear in corners of the FF sensor (and FF means 24x36mm, and that size comes from Leica-format film, which is considered as some sort od standard (not really relevant today really))
You are already achieving almost maximum with what you have.
Perhaps what you need is to use smaller aperture (higher f-number) to limit CA and coma (that also means longer exposures).
I have very good results with Canon FD 100mm f/2.8 lens, but I had to stop it dowm to f/3.5 ~ f/4 to remove CA (blue/red collars around bright stars).
(see
here)
I had best results with 200mm -400mm prime lenses with external aperture mask (made of black paper or some sheet material) placed in front of the lens.
This has the same effect as increasing f-number (F/4 ~ f/5.6) but there are no diffraction steaks (because the aperture is round, not segmented as with internal iris).