I'd buy a Minolta - either the II or the IIIF, depending on how important the extra flash functions were. Personally, I'd be happy with the II. If you know from experience how your flash equipment compares with daylight, you only need a continuous meter to balance flash and continuous - only the continuous is unknown. I know what I'm trying to say, but I'm not saying it very well.
Accuracy etc: The short answer is that the three meters are remarkably close. I'll go into more detail in a minute.
I've never felt the need to use a colour meter to balance strobe to strobe, but I don't have enough experience with different brands of strobes. My only use of a colour meter with strobe is to balance to ambient, in which case you can usually manage without measuring the strobe itself. I think that my situation is very similar to yours. Perhaps someone with more experience with colour meters and strobes could answer that one.
Back to accuracy.
I think that colour meters are fairly consistent, especially between 5500 K and 3000 K. My MII drifts a bit below 3000 K, but it is old. The maximum difference between any of them is about 12 mireds in the 2300 K to 2600 K region. That's not a big difference in practice. The green-magenta index is comparable as well, although the fact that the Gossen reads in steps of five does lead to differences. My MII consistently reads -2 for blackbody sources, but if I take that into account it compares well with the MIIIF.
When you think about how the readings from a colour meter will be used, the differences between the meter readings look unimportant. If you are balancing sources to each other then absolute accuracy isn't necessary, of course. All the meters show good precision, equalling the resolution of their displays.
For balancing a source to a film type accuracy is more important, but I haven't noticed enough discrepancy for it to be a problem.
Here's a summary of the nominal shift values of the least dense standard filters.
Eighth CTO, CTS: +20 mireds
Eighth CTB: -12 mireds
Eighth plusgreen: CC 3.5 G (CC035G)
Eighth minusgreen: CC 3.5 M (CC035M)
Wratten 81: +9 mireds
Wratten 82: -10 mireds
KR 1.5: +15 mireds
KB 1.5: -15 mireds
CC green: CC 2.5 G (CC025G)
CC magenta: CC 2.5 M (CC025M)
(I don't have the need to use camera CC filters in 2.5 unit steps)
Those values should give some perspective on the degree of accuracy (or lack of it) required from the meter. I think that all the meters I've used have been sufficiently accurate.