An analysis of a parallel-plate UO2-fueled Fission Electric Cell is developed that includes a detailed treatment of the fission-fragment initial-energy spectrum, energy-charge loss during slowing, and energy dependence of the total range. The treatment of fragment transport is based, as much as possible, on correlations of experimental data. However, available data are skimpy, and several discrepancies, e.g., between available differential and integral energy-loss data, are noted. The importance of an accurate fragment transport model is demonstrated by the differences in efficiencies obtained from this detailed treatment, as opposed to earlier calculations that used simpler models, e.g., relative differences between models of as much as 15 and 80% are attributed to the treatment of the fragment charge and energy loss, respectively. The calculations are also shown to be fairly sensitive to the total-range-mass correlation, but only weakly dependent on the choice of the initial fragment charge. While efficiencies for the parallel-plate cell with reasonable fuel-layer thickness are found to range from 2 to 10%, efficiencies for cylindrical or spherical geometry may be 5 to 6 times this, and the concept may be competitive for certain specialized applications.