Molecular and macromolecular high-permittivity organic gate dielectric materials have been the focus of recent experimental research as a consequence of their promising properties for organic and inorganic field effect transistor (FET) applications. Two types of molecular thin films, self-assembled nanodielectrics (SANDs) and cross-linked polymer blends (CPBs), have been shown experimentally to afford high capacitances and low FET operating voltages. In an effort to design optimized nanostructures having even larger capacitances, lower leakage current densities, and further reduced FET operating voltages, we discuss approaches for computing the effective permittivities of each nanodielectric motif and investigate how molecular arrangements impact overall device capacitance. The calculated frequency-dependent capacitances, derived from Maxwell-Wagner theory applied to the Maxwell-Garnett effective medium approximation, agree fairly well with the experimental values for the two types of nanodielectrics. Predictions of larger capacitance SANDs are made with the two-capacitors-in-series equivalent circuit, where the layered, self-assembled structure is viewed as two different capacitors. The Maxwell-Garnett and Polder-Van Santen effective medium approximations are used to predict the dielectric response of higher permittivity polymer cross-linked blends. In calculations showing good agreement between theory and experiment, and with all parameters being equal, it is found that greater capacitances should be achievable with cross-linked composites than with layered composites.