The development of ceramic-ceramic composite nuclear fuels benefits from thermal modeling by providing an understanding on how fabrication variables, such as phase fractions, densities, and geometry, will determine effective thermal conductivity. Finite element method (FEM) two and three dimensional programs were used to predict the thermal conductivity of composite UO2-BeO materials. The FEM modeling results were compared to the measured UO2-BeO fuel sample thermal conductivities. The comparison showed that the thermal modeling was in good agreement with the measured values. These benchmarking cases with the FEM thermal modeling method successfully demonstrated the potential of the models to accurately predict the effective thermal conductivity of an enhanced thermal conductivity oxide nuclear fuel. The FEM thermal modeling was used to predict UO2-BeO nuclear fuel thermal conductivities with different BeO percentages, and then the reactor fuel thermal behavior was analyzed using the UO2-BeO nuclear fuel thermal conductivities and other material properties. The analysis results show significant temperature decrease for the UO2-BeO nuclear fuel compared to the traditional UO2 fuel, and then the safety of the reactor would be improved.