I have had a paper accepted for a conference this summer, which presents the preliminary results of my PhD research. However, I'd like to present my final results at another conference in the fall, the deadline for which is in a few weeks, and which requires a full paper submission. My question is: would it be unethical to submit the first paper (in a re-structured form) in the hope that it will get accepted for the later conference, and then subsequently update the results before the final deadline?
"May I submit a paper to another conference that is essentially the same as a paper that is already published or accepted for publication?"
"May I substantially change the content of an accepted conference paper after peer review?"
* Unless it is disclosed to, and permitted by, the PC/editor.
To add to @ff524's absolutely correct answer, you can still submit to that second conference. All you need to do is cite your first paper, state that this paper only adds results A,B,C. It is possible that the modest additions to your first paper are enough to merit publication on their own; however you must be honest and let the editors/referees decide this.
In short, you may submit your updated paper, provided you are completely honest about its differences with the previously published work. If indeed those differences are trivial (as suggested by the title of the question), then obviously this is pointless.
ff524's answer is true for the general case. There are two exceptional cases where you will be allowed, or actually required, to make substantial changes to the content. Note that none of them will happen because you want to add a bit more.
First, your paper might be put into a gatekeeping process. In this case, the acceptance is conditional on making certain changes and the reviewers being happy about these changes at a later reviewing stage.
Gatekeeping only occurs when the program committee requests it; you can't trigger it (and don't want to). It only allows you to make the changes the reviewers requested from you (normally to provide better argumentation for your conclusions, or to describe your empirical methods in your detail), not to include any results you forgot about or didn't have at the beginning.
The second case is if you find an error in your research which renders your paper wrong. In this case, you should contact the program committee, inform them of the error, and offer to retract the paper. If you have the improved results, and they are not completely changing the main statement of the paper, the program committee may allow you to keep the paper in the running, but will require you to rewrite the parts based on the mistaken results.
It is completely unethical to substantially change a conference paper after acceptance, just as it would be completely unethical for a used car salesman to substitute a different car after you'd agreed to buy. The programme committee accepted the first version, based on the advice of the reviewers; they did not even see the second version. If the same thing happened in a journal version, the second version would be sent back to the referees but there is no second round of reviewing for conferences.
Honestly, I'm worried that you're even asking this. It suggests that, in your mind, the primary goal has become publication, rather than doing good research.