In all honesty, I think that you are setting yourself up for a failure. But if you still do believe in yourself, you definitely should chop your project into smaller pieces, and work on one piece at a time.
-For example you could create one type of a game, where you can just put trees or structures down on a map, and later on evolve that to a world editor.
-Secondly you could do a small game like pong that works as a 2 player game over a network. Eventually you could learn about how to create a multiplayer game and apply these things into your online game.
Surely these should be split even in smaller tasks, but the idea is that like jigsaw puzzle you start putting it together from edges piece by a piece and maybe create some islands of area then merge some islands together and eventually you'll have your whole picture finished.
Also when it comes to engines, you could just toss the coin to determine if you should use Unity or Unreal, other wise you might end up spending next 1 to 2 years pondering if Unity is better or if Unreal would the best choice. Just choose one and stick with it. Although I do personally think that Unity is more suited for low poly things, since Unreal doesn't seem to be optimized into low polygon things.