The oldest one is from Feb 14, 2016 but I'm sure historically significant changes to the code were done since.
Indeed, date opened is not always a good indication of controversy or significance, it could also mean a change which has received very little review due to its insignificance, or very little rebasing/nit-fixing due to lack of time by the creator. It's common for more important changes to be the focus of more urgent review allowing them to be merged sooner. Perhaps you should try sorting by number of comments or number of reactions, but even then it depends what you deem "significant".
Does any one constantly (like everyday) review all commits outside of the bitcoin core team?
By definition, someone who is actively involved in the review of PRs would be considered part of the "team". There is really no team, just volunteers, anyone is free to come and help out on GitHub, no requirements or prerequisites. Many users help out in this way, and you could be one of them :)