Everything varies for everyone, but I think it's pretty objectively hard to fit your own campaign into real-world history or a well-established fictional setting -- especially if any of your players know the setting better than you do.
Subjectively hard, but not objectively. A lot depends on how willing one is to strongly assert that one's campaign version of the world the players know is not the same as the one in the book/movie/etc.
Certainly it's easier to write swords & sorcery fiction in a made-up world full of quasi-Vikings, quasi-Corsairs, etc., than to meticulously research real-life history or meticulously create (and convey) elaborate society from scratch.
Some people enjoy that kind of meticulous research. Those who don't, shouldn't bother trying to run such a game or writing such books.
In addition, my efforts have never been to create a Middle-earth setting or game rules for it that is easily accessible to casual or non-readers of the books. But at least one reading, certainly two, should give one enough background to run such a game. Same goes for historical settings; I don't think you need to do enough research to create a Master's thesis to run such a game.
This is also why I emphasize the Fourth Age as a good time for a Middle-earth campaign. You really don't need to know much more than what a reading of the books or a viewing of the movies would reveal. Need to know more about the history of a place? Well, there are now tons of resources, printed and online, where solid synopses of just about anything are right at your fingertips, that can give you more than enough info in five minutes of reading.
Exactly. Those who've read the series twice aren't in the same boat.
Twice is more than enough. Once will do. No one expects a doctorate-level presentation. It's just a game.
Certainly, some D&Disms are bound to creep in, if only because the RPG genre has its own demands, but so many D&Disms seem superfluous. So, introducing hit points and then finding out that you need easy healing magic is understandable, but creating a mace-wielding crusader as the only source of that healing is just weird.
I don't think you'd need a separate supplement at all. You'd just need base classes and rules that allowed you to play characters like Robin Hood (or Faramir, or Legolas), the Knights of the Round Table (or the Knights of Gondor), Conan (or Fafhrd), etc. You know, Rangers without (mandatory) spell-casting, Fighters with some skills, etc.
You
do need a separate supplement. People - including me - have indicated that going back to D&D's roots as a game is desirable. People don't want radical revisions of the classes and rules; that's been tried, and that's why Next is in playtest, and why it harkens back to past editions. D&D, for better or worse, is D&D, and many have indicated that that is what they're looking for. So, yeah, I assert that what you want has to be done as a separate supplement, because it would involve the kinds of changes that have proven to be contentious among D&D players. The Next playtest has been getting consistently positive feedback and discussion, enough that it would be unwise to veer off and make the kinds of changes you seem to want.
Of course, if you feel that what you've seen in the Next playtest is moving the way you like, then that's another matter.