[D] Is there a more systematic way of choosing the layers or how deep the architecture goes when creating a neural network?

So I'm learning about deep learning and neural networks and I'm really a bit confused on this part. I'm generally familiar with the layers available and how they work (at least those that are widely used) But I'm still having a hard time trying to figure out what to use on what. Is there a more logical or a systematic way of doing this? like mathematically or something? I'm down for experimenting but I'm just trying to avoid the rabbit …

architecture deep learning least machinelearning network networks neural network neural networks part work