That list is not "scientifically proven" since the sample size is far too small (only 6 lists) and therefore suffers from selection bias (since it only represents a small fraction of the lists out there).
Here's a far more "scientifically proven" (or rather, "statistically proven") list based on weighted rankings using a much larger sample size of 54 different lists:
This list seems much more accurate to me, though Goldeneye still sticks out. Not just because I don't agree with it, but because I don't feel like it really holds that sort of place in popular opinion.
I agree with most of what you're saying, aside from maybe the length comment.
But I think this is partly happening because their video coverage is in a transitional period. Once they figure out exactly how they want to do reviews, which from what Jeff has been saying may be soon, I suspect quick looks will find their proper place again.