Whew! A mod. Man, even your scaled back version of your old idea was WAY too much. I was gonna say to start with something very small, like a puzzle game. Even a 2D platformer could be too big of a task for most people as their initial finished game.
But yeah, a mod should do nicely. Even that can easily get way too big though. Like you were saying about filtering the sound effects to lose treble as they move into the distance, if that's not already a capability of the engine, forget it. Just work on solid design within the game engine's limitations.
Even though it's greatly scaled down from before, your current Neverwinter Nights mod list of goals worries me.
- Good looking "shores"
- Range of weather effects (including sun cycles if possible)
- Realistic sound (further away looses treble etc)
What is your overall encompassing goal? Is it to make the game look and feel realistic? I suggest that you give people a reason to play based on interaction, not piling on special effects and levels of realism. Relying on that kind of stuff will require more of you than you'll ever know, and being the small project it is, won't capture people's interests NEARLY as much as focusing on interaction. Seriously, putting effort into realism is going to net you maybe 5 to 10% of the player interest that the SAME LEVEL OF EFFORT put into interaction will net you. And some can argue that it may not net you any fun at all.
And even in interaction, focusing on REALISTIC interaction could also waste a lot of time and energy. Go for interesting interaction. The type that engages the mind, or dexterity of the player, or both. Work smart in that sense. The more concise your design, the better. You can come up with a smart design that doesn't require tons of content/programming based on mimicking the environment and the human mind. When designing games, it's so easy to just pile on more and more. "Like this game everyone likes, but MORE". If you really want to impress, you'll have to step up your design skills, and build something concise and sharp. Unfortunately, that kind of skill requires practice. Several games of practice. However, I'm just one voice. Definitely take my advice with a grain of salt. But do look for themes across all of the advice given to you.
I think you're heading in a great direction by simplifying. The choice to turn it into a mod is a great compromise between your original idea and what people here are saying. I commend you for doing that. I've seen many with ideas like your original concept too bullheaded to take the years of experience the people on this forum and others like it have to offer. Most people here can tell you what fate was suffered. It can be a romantic idea, "doing what they all said was impossible". Just keep in mind to work smart. Simplify. It requires a lot of skill to do it well and keep the player engaged, but the payoff is pretty big. You may indeed accomplish the "impossible". Just stay smart about the steps you take.
Good luck.
edit: of course it wasn't his intention, but this quote is quite pertinent to game design, I feel:
"Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction."
- Albert Einstein