You might find it far easier to do this sort of thing if you sketch with pencil on paper. Doing it with a tablet is much, much harder than drawing with a pencil - a lot of artists don't even bother, they use scanned sketches for digital painting.
I think you misunderstand Loomis's point, though. His "mannikin" is an abstraction of anatomy that can be useful to construct a figure from imagination, but is primarily intended to help you parse what you are seeing in an actual human model. As you say, anatomy is intricate, so it helps to have a handy breakdown of it into simpler blocks; that's the mannikin.
You can start out with a contrived mannikin and gradually adjust it into a more realistic shape, but if you are just starting out, why not start with the realistic shape and distill that into the mannikin instead? Work on learning the real figure - as long as you are understanding what you see, you'll be able to convert that knowledge into skill in drawing without a model.
There are tons of ways to simplify the figure for beginning a drawing, from structural Loomis to gestural Vilppu. Find your own way. Later on you'll likely find that you are focusing on different things in different models and poses, so your "mannikin" is different every time. Focus on the things that are important for a particular drawing. It's about capturing the human being's motion, not about formulas!