I agree for the most part. The problem with reserving 95-99 ratings for "hall of famers" is obvious - none of us know for certain if a player is getting into to the hall of fame until either they are well into their already hall of fame worthy career, or until they are basically at the end of their career and have met certain milestones and criteria that seem to be important to hall of fame voters. Even then, we may not actually find out if they are "hall of famers" until 2-3, sometimes 4 decades later when they actually get voted in.
Therefore, a player early in their career, playing at an elite level will actually be underrated using the 95+ equals hall of fame only criteria. This makes no sense when you consider that many of the truly greatest iconic players in NFL history were pretty much great right out of the box (i.e. Jim Brown, Lawrence Taylor, Ronnie Lott, Reggie White, Joe Greene, Anthony Munoz, Dick Lane etc) while others shone brightest in their first 2-4 years in the league (Earl Campbell, Gale Sayers, John Mackey, Mike Ditka, Marcus Allen, Eric Dickerson etc). In fact, the career arc for many NFLers is to burn brightest early in their career and then either fade away out quickly, or carve out a niche as a solid pro later in their careers. This seems particularly true when considering a position like running back where one or two injuries and wear and tear can knock a former beast to average back in a matter of a season or two.
The bottom line is ratings should be fairly heavily weighted to how the player is performing at this moment, in this season. I mean an elite season is an elite season because it can be some what easily compared to player performance before it. If a player is playing at an elite level, and is not just the product of a system, this should be recognised in their ratings.