The most reliable method I have for weeding out ideas is to try to use them. If you have time to obsess, you have time to write, right? So do it. Pretend you've already decided that they're in, and make a serious attempt to fit them into the story.
75% of the time, the motivation for using a particular idea will fade as soon as you start actually having to work on it. Some things seem more fun in your head than they are in real life...and that's okay. One less thing you have to worry about.
20% of the time, you'll hit your stride with an idea and really get into writing about it...only to read over it later and find it will do more harm than good for the story (e.g. takes too much time, distorts the focus of the narrative, leads the plot towards a path it won't actually go down, etc.).
Or, in many cases, it will just end up being an essay full of idea-vomit that you didn't notice because you were having too much fun at the time. ^^; This is why it's important to take breaks and come back to a script/draft with fresh eyes, over and over again...If you like what you wrote despite its uselessness in-story, you can always save it as reference material. Otherwise, at least you got it out of your system.
The remaining 5% of the time, you'll actually have a good idea that could do good things for the story. From there it's just a matter of editing, and watching tone: often the main difference between a set of equally useful ideas will be the emotional energy they hold. From deciding whether or not to have a character make a sarcastic comment during a fight scene...to deciding whether or not the victim the hero is trying to save will live or die during the attempt. Both paths can be interesting and have value, but they place different kinds of expectations on the story/character.
You could decide you want the character to be 100% serious, and the story to be unflinchingly hopeful, and eliminate the other paths right off the bat. Or you could decide that the character's seriousness comes with attitude, and that you can still write a hopeful ending-- but with more emotional weight-- if the victim dies, and weave these new options into the existing framework of the story. Or you could decide that these new ideas lead to more new ideas that you like better than what you have, and continue the process.
All of that depends on knowing what kind of story you want to tell, though. If you're still on the fence about how the story should 'feel' overall (like whether a plunge into death and darkness could even fit into it or not) waffling about the smaller details isn't going to accomplish anything. You can't build without a foundation.