As is known, atomic and very small molecular systems can be realistically simulated by quantum mechanical models. In complex chemical systems, however, the natural parameters are not only electronic density and energy but also entropy, temperature, and time. To approach this description we use a hierarchy of theoretical models. In model I , the system is considered as an ensemble of fixed nuclei and electrons-the standard quantum chemical approach. However. our definition of model I is broader, since it includes aspects of solid-state physics. In model 2, the system is considered as an ensemble of atoms (or ions) and atom-pair potentials are obtained using data from model 1. In model 3, the phase space is scanned either for the generalized coordinates (Monte Carlo) or for both space and momentum coordinates (molecular dynamics). In model 4 (presently not considered) the fluid dynamical equations are solved making use of coefficients and parameters obtained from the previous models. Both theoretical and computational improvements are needed at each level in order to reach a sufficiently realistic simulation for complex systems. We have summarized some recent progress obtained for models 1 and 2 related to new methods for molecular computations and studies on three-body effects and energy band computations in DNA-related polymers. We have considered as examples of a complex chemical system the structure of water surrounding DNA (with counterions) and enzymes. Our results from model 3 include the first determination of the position of the Li+, Na+, and K+ counterions in B and Z DNA at room temperature at high relative humidity, and hydration studies on enzymes including variations due to the solvent pH.