Learn one subject after another or all in parallel (like at an university)?

There are a lot of things I want to learn. It's really hard to prioritize them and even if I do - a lot is left to undefined time to the future.

What I've realized is that at an university we used to learn everything in parallel, in small steps.

What if I apply this approach in my self learning too?

Would it be too slow for fast changing fields (ML, Big Data for instance)?

I am gonna give it a try anyways, just wondering if someone else here has similar experience.