There is a vast set of industrially-relevant materials for which traditional molecular modeling techniques are not sufficient. Semiflexible, anisotropic, and stimuli-responsive materials all present distinct challenges toward the traditional techniques of developing theories and models with which to describe them. In this work, we develop specialized models to handle several different types of difficult materials, and then we use these newly developed models to perform a study on them. First we examine conjugated materials used in organic photovoltaics by creating an anisotropic coarse-grained model capable of describing the necessary twisted and fused geometries. A network analysis is performed to determine that molecules that are the most twisted can resist crystalline ordering and tend to work best for forming morphologies that are most fully connected and can easily transmit excitons. Second, we examine a semiflexible polymer brush that experiences nematic interactions by extending the theories that are commonly used to describe brushes with a wormlike chain backbone and Maier-Saupe pair interactions. We then examine the effect that grafting density and polymer length has on the nematic ordering that the brush experiences, and find that increasing grafting density tends to increase nematic correlation lengths, while increasing polymer length increases configurational entropy and decreases nematic ordering. Third, we examine polymers grafted onto cellulose nanocrystals under various solvent conditions by using a coarse-grained model with a three-body potential to implicitly model different solvent effects. We derive free energies as a function of distance and orientation between pairs of cellulose nanocrystals and use them to parameterize a coarser model and examine morphologies of cellulose nanocrystal suspensions for water percolation. Finally, we examine the self-assembly behavior of multi-component bottlebrush polymers in two architectures, diblock and Janus bottlebrushes, by extending a polymer model to include different architectures. We find that the two architectures have very different phase behavior, diblocks behave similar to linear polymers while Janus bottlebrushes rely more on the extent of side chain stretching to dictate their behavior.