Tony,
Terry is much more knowledgeable than I with scanning, so I know he will have some sharp data for us, but just from my experience, the main problem with scanning targets or anything that will be used as a benchmark is the scanner's light source/color temperature and metamerism.
That's the main reason Mike recommended the Canon LiDE scanner because of the "cold light" source.

So, off the top of my head, sounds like opening a can of worms with no expectation of good results.

On the other hand, I get the feeling that you are not using the editor in correct fashion as it will allow tweaking your profile with super results.
For example: you can tweak flesh tone, without effect on trees, and tweak the green/yellow foliage without any effect on flesh tone.

The procedure is to match the right side of the editor (which has two screens) to the print you made of a standard image (Digital Dog or whatever you like) by adjusting sliders of the various controls until you make the right side screen match the print.
Then you make your corrected profile, and make another print. Let it dry a bit and cure the colors, and now compare it to the reset right side which matches the left side.
It should be very close, another final tweak; maybe a touch less saturation or more brightness?   You still do this by making the right side match the latest print; not by moving sliders the wrong way.
Last hint is to make prints, and subsequent prints. Softproof is not a substitute for a real test print.

Maybe this helps??

Fred

I'm not sure that it will and could confuse the matter. Would you allow Vuescan to modify the image or use the raw version? I wouldn't know. I find that scans from a print are generally lower in contrast than the original so that will affect RGB values in addition to the scanning process modifying colours.
After all that, I cannot see how that data would help within PP.
Remember that in the PP editor, the image on the RHS has to be made to look like the print so it's a sort of reverse process. Therefore using data from a scan would be difficult to interpret.
The PP editor has a colour circle (I recommend using the wizard dropper) so that shows you where a particular colour is on the circle. The editor is very much a visual editor but if you want to read the RGB values in PP, use a tool like Color Cop to do that; that will give you the numerical balance of RGB values but I'm not sure how useful that really is.
That come with practice but a good understanding of the physics of light & colour may help.
The beauty of the PP visual editor is that the latter knowledge is not essential and trial & error can be used - as long as the RHS ends up looking like the print.
Don't forget to illuminate the print with either good daylight or a proper daylight source like an Ott-lite.
Terry

Be careful, artists have a different view of colour and what colour is made when they (paints)are mixed. It is not the same as light.
Artists tend to think, for example, that yellow is a primary colour, it's not. Yellow light is made from Red & Green.

I have not used PP for a while for a printer profile so just for fun I resurrected some scanned targets fro PP and re-made a profile. This was for Ilford Smooth Heavy Weight matt. Except for some saturation increase needed on the print, colours were good and greys perfectly neutral. I had the bear in mind that a matt paper does not produce particularly good contrast or saturation.
Terry

Further information re:
Again, just for fun  I made 2 prints on an A4 page of 4 images, including "real world" images. One was with my PP profile and the other with a "proper" custom profile made my my local colour management company with their top quality, Multi £'000 equipment.
Even I was surprised, the PP profile was just as good, the prints had more "bite" and excellent colour. I could probably afford to reduce the contrast a little, that's all.