I'll admit that I'm a little dubious of the survey-based approach to design, and more so of the implication that there's a single--or even a dominant--design for a "great" RPG: I think that different people like different things in an RPG, and a game based on aggregate responses won't necessarily please more players than one based on a non-aggregate design. For a few examples: some people like story-heavy RPGs, while others prefer action; some want a lot of statistics to min-max, while others prefer to avoid that; some want an open world, while some prefer a more directed experience.
If I may, I'm inclined to suggest simply making the RPG that you'd like to play: it seems likely to me that if it's something that you'd enjoy, then there's a good chance that there will be others that will like it, and that you're more likely to remain invested in a project that appeals to your own tastes.
If you want feedback on your ideas (which is a good idea in general, I do believe!), I'm more inclined to recommend coming up with an idea and then posting in the Game Design sub-forum for feedback specific to that idea. (Even better if you have a prototype to show!)
