I don't know if you're into math or programming, but if so, then in addition to Chaqa's suggestion and 2WL's VERY important point, in an argumentative paper you can kind of think of the large-scale structure in a recursive way. You have a large-scale argument:
We want to show X.
1. Premise A is true.
2. Premise B is true.
3. Premise C is true.
4. Premise D is true.
5. Premises A, B, C, and D imply X.
So X.
For each place I've numbered, you recursively bubble down and give an argument (maybe of the same structure) for A, B, C, D, or (in the case of 5) why A-D imply X.
It also would be a good idea, once you're done, to go back to the beginning and give an outline of the structure of your argument so that people understand why you're getting into these side-things.
This is, as I say, for argumentative pieces, i.e., things defending a thesis. I've seen the structure used, for example, in analytic philosophy. Sometimes you can mix it up and be a little less formal at times, and kind of invite your reader along on open-ended exploration with you, but you should start with a structure like this as your baseline.