What are the observed issues with the stock design?
Straight out of the box, the overwhelming problem is that it will self-destruct on the first minor crash, that's a fatal flaw (hence my recommendation...). Presuming that you rebuild it out of balsa to avoid that, the performance issues are (in no particular order):
1 Rate-limited corner
2 difficult tracking in level flight and straight-line segments
3 poor response to turbulence
4 hops out of corners
With no other changes I would expect nothing you can do to the wing will fix #1. 2, 3, and 4, will likely alter it but you have to separate out the effects of your changes from other items that will likely affect them far more. I can't entirely follow what you are proposing as a change aside from the swoopy wingtips but if involved raising the aspect ratio AND introducing taper or the equivalent, those will tend to offset each other for 3 and 4.
I would make 3 changes starting from my all-balsa model as a baseline:
First shorten the tail moment (arm, there you go Howard, I do talk good English) by around an inch, and enlarge the area to keep the same TVC. Intent is to solve #1. I would make that change without any other changes to aid in evaluation.
Second would be to keep the same aspect ratio, but add taper, maybe 3/4" more root and 3/4" less tip chord. I would also make that change by itself. The goal is to reduce the tendency to bounce around in roll/yaw in turbulence.
Third, providing that I got the tail moment right in step one, I would be tempted to reduce the aspect ratio to soften up the response to elevator motion.
That would take a long time to get through. Note that I am certainly not going to do any of these changes. If I was going to pursue this line of reasoning, I would be far more inclined to start with a Medic or Doctor instead of a Skyray since I like the airfoil better and it's a lot closer to the end goal already. The Doctor, in particular, has only one and a half of the above-mentioned issues and I could be competitive in any contest in the world with it, as long as the air was perfect.
I don't see it as a candidate for a better plane, but rather a tool to answer a "what if?" question.
The problem with doing this is that, for sure, the airplanes will fly differently. Whether it flies differently as a result of the changes you intended, or differently as a result of some other difference (like alignment, etc) is another issue. I would suggest 5 of them, compared to 5 Skyrays. Trim them all the different ways you can think of. Have them and have them evaluated by multiple experienced pilots. Compare notes, find the common attributes and limitations, then you will have some idea what the effects of the change you made are and how much they changed. One example will be extremely prone to unit to unit variation, and with any but the most experienced pilots, you will also tend to blur variations in the airplane performance with difficulties you might have with the piloting.
Not to mention that the characteristics you are intending to experiment (like the shape of the lift distribution and induced drag) are such tiny effect at the best of the time, and also utterly swamped by tiny moment-to-moment changes in the way the engine runs.
It boils down to this - there have probably been 100,000 Green Box Noblers built since 1957. No two of them fly the same, and it takes a while and multiple examples to figure out what the design characteristics are, and how to adjust them VS someone putting one extra coat of Ambroid on the rib/spar joints.
Brett