Polymer nanocomposites (PNC) are an exciting class of materials with a wide array of applications. Whether the desired application involves isotropic or anisotropic polymer and particle structures, controlling aggregation and/or anisotropy in PNCs remains crucial to engineering composites with desired properties. Many methods have been developed to address these challenges. Three methods of interest include 1) the inclusion of anisotropic particles in the PNC, 2) block copolymers used as anisotropic templates to pattern particles, and 3) particle surface chemistry modification to increase dispersion. Significant experimental progress has been made in all three of these methods, but deeper fundamental understanding in each of these domains is necessary to continue to improve experimental control over PNCs. Efficient simulation methods like Polymer Nanocomposite Field Theory (PNC-FT) and Theoretically Informed Langevin Dynamics (TILD) are a powerful way to gain these insights when used alongside experiments. With regards to anisotropic particles, PNC-FT was used to gain a better understanding of an experimentally observed phenomenon where gold nanorods (NRs) were observed bridging cylindrical domains in a diblock copolymer film. The simulations supported the idea that the observed orientations were indeed energetically favorable and not just kinetically trapped, and found ranges of NR length where bridging is expected to be most favorable. With regards to using block copolymers as anisotropic templates, in addition to the bridging nanorod study just described, PNC-FT was used to investigate the effect of diblock domain deformation on particle-particle interactions when particles are confined within a diblock domain. It was found that for an A-B diblock copolymer nanocomposite, domain deformation can lead to stronger particle-particle attraction compared to interactions within a homopolymer melt, with attraction strength and separation distance tunable by properties like grafted chain length and density, particle size, and diblock A-B repulsion strength. Finally, with regards to surface chemistry modification to increase dispersion, TILD was used to better understand particle clustering behavior as a function of particle volume fraction and particle-polymer attraction strength. It was discovered that particle cluster size distribution and cluster structure change relatively slowly over the sharp phase boundaries present in phase diagrams.