Did you forget how many publications are open access and/or available as preprints?

A huge fraction of science already is free. Yes, there are downsides--for instance, preprints aren't peer-reviewed, and open-access publications have to pay for their editing work somehow so they charge publication fees (which for research that is otherwise very inexpensive can be an unwelcome type of gatekeeping).

But in many critical fields, it's the norm. Out of curiosity, I went through a bunch of papers in the latest IPCC report (I forget which chapter I picked). I looked at twenty before I got bored--every single one was open access.

So, the bottom line is that Elbakyan is a bit late to the party. Yes, you can make a utilitarian argument that Sci-Hub is extremely valuable. (As just one line to dispute it, you can make a counterargument that you can email researchers asking for their paper and usually get it. You can make counterarguments to that, etc..) But compared to the Open Access movement, it's small fry. Physics is almost entirely on arXiv, and a lot of CS and engineering and such are too.

So Elbakyan has--for this kind of contribution--extremely stiff competition from the likes of Gisparg (arXiv); Giles, Bollacker, and Lawrence (Citeseer); Varmus (PubMed); Holdren (one-year-until-open rule for U.S. funding--unless we want to give Obama the credit here) and so on. Most of the most impactful and relevant research is now open access. That which is not is still important, but it seems like you're not even aware that this enormous change has already taken place, and the cases where critical papers are behind paywalls (especially for over a year) is already only a modest fraction of the total.

So, impactful? Yes. The best ever? No, not even in the indirect sense that you're enabling here.