In the present paper, a novel and efficient model was developed for predicting the effective thermal conductivity of the composite materials at different filler percentages. By introducing the relative radius as a parameter, the effective thermal conductivity can be predicted precisely when the thermal properties of filler and matrix are prescribed. The model employed the resistor network strategy to achieve a highly efficient prediction during the overall conductivity calculation. To verify this model, a wide range of two-phase composites, including Cu/PP, AlN/PI and a self-developed thermal conductive adhesive, were analyzed. The model-based simulation values showed good agreement with the experimental results. Moreover, a discussion on the effects of the newlyintroduced parameter was given. Finally, the relationship between the filler radius variation and the thermal conductivity of the composite was studied.