Some great points here and super well said!!!!! This does an excellent job of summarizing some of the best advice I've received so far in my MSTP.I think everyone has given good advice already but just to add. You need to do activities in which you input time and get help that is out of proportion with the time invested. The easiest of which is to simply decide what lab you want to definitely join when you go to graduate school. By doing that you will already be able to shave off time and will not hinder your medical school endeavors. You may think this is easy but too many people are wishy-washy and unable to commit.
You basically look at all the faculty at your school, list all the ones of potential interest, keep the ones that have a good track record (never pick one who has a bad track record but you think you will be the special one. You won't.) Then really try to assess and figure out who is likely to be there for 4-5 years. The way to judge this is how long they have already been there, what kind of work they do (how tied down are they), and you can just ask them straight up. I bet your final list of faculty with a good track record won't be too long so then you can just schedule meetings and talk to them. Once you find the one you like you can just plan out a research area (this is also where the above posters advice about being open in communication is important). You may say, "Oh but no one with my interests falls in the list." My response is that you should expand your interests or just be okay taking the risk. People that can help you assess the risk: MSTP office and older students.
Getting into the weeds of your project is an example of an activity where your time investment is NOT worth it. There are stories of very exceptional people doing work that they did in undergrad and basically buying all the equipment they needed to get everything ready in the lab for them to join. To me that sounds boring because if you already knew everything about your work you are not growing. Now doing some minimum work in the lab just to understand how things work, maybe learn a specific method and help generate data is possible (<5 hours/week) but not necessary. You could also just get this with one effective rotation if your school does summer rotations. 20 hours a week is not worth it because it is not enough time to build your own dissertation and too much time to be a contributor.
Focus on learning med school stuff and determining what specialty you want to do. These are much more useful activities and will give better returns. If you screw up Step 1 and want to do certain fields nobody will care what else you do. If you decide what specialty you want to do early on, you can easily make yourself more competitive for that specialty during PhD time.
Seek out advice of older students who accomplished what you want to accomplish. Your MST program should have people like that. TO do that first figure out what you want. It sounds like time to degree is important but make sure you balance all factors and see where they fall (time to degree, big paper, lots of papers, etc).