Molecular machines moving along DNA or RNA are involved in almost every aspect of genetic processes. Since these machines work across time scales of several orders of magnitude, it is hard to study them computationally from a single approach. In this talk, I will introduce my modeling and computational work on DNA/RNA helicases, polymerases, and a packaging motor, by using all-atom molecular dynamics simulations, coarse-grained models, and stochastic or kinetic modeling based on experimental data. In particular, I want to show some interesting perspectives obtained by studying simple models related closely to experimental results. I will explore the possibility of using these models 'top down' to guide detailed dynamics simulations 'bottom up' in order to discover functional mechanisms of these molecular machines.