In head computed tomography, radiation upon the eye lens (as an organ with high radiosensitivity) may cause lenticular opacity and cataracts. Therefore, quantitative dose assessment due to exposure of the eye lens and surrounding tissue is a matter of concern. For this purpose, an accurate eye model with realistic geometry and shape, in which different eye substructures are considered, is needed. To calculate the absorbed radiation dose of visual organs during head computed tomography scans, in this study, an existing sophisticated eye model was inserted at the related location in the head of the reference adult male phantom recommended by the International Commission on Radiological Protection (ICRP). Then absorbed doses and distributions of energy deposition in different parts of this eye model were calculated and compared with those based on a previous simple eye model. All calculations were done using the Monte Carlo code MCNP4C for tube voltages of 80, 100, 120 and 140 kVp. In spite of the similarity of total dose to the eye lens for both eye models, the dose delivered to the sensitive zone, which plays an important role in the induction of cataracts, was on average 3% higher for the sophisticated model as compared to the simple model. By increasing the tube voltage, differences between the total dose to the eye lens between the two phantoms decrease to 1%. Due to this level of agreement, use of the sophisticated eye model for patient dosimetry is not necessary. However, it still helps for an estimation of doses received by different eye substructures separately.