Advice from Hal Varian’s monograph, How to Build an Economic Model in Your Spare Time [pdf], most of which applies to things not related to economics:

1. Look for ideas in the world, not in the journals.
2. First make your model as simple as possible, then generalize it.
3. Look at the literature later, not sooner.
4. Model your paper after your seminar.
5. Stop when you’ve made your point.