An improved solar parabolic trough collector model has been developed in this study, to overcome the weakness of the previous published studies in predicting accurate performance, at higher operating temperatures. To check the accuracy of the proposed model, both on-sun and off-sun tests have been considered. The on-sun tests are aimed to check the potential of the model for predicting the thermal efficiency while off-sun tests are for heat losses. The results have indicated that the efficiency uncertainties of the present model are lower than those obtained during the experimental tests of Sandia National Laboratory for all the cases. When compared with previous published models, the proposed model is ranked first, more accurately than those of NERL, Garcia, and Behar. An average uncertainty of less than 0.50% has been obtained while Behar's model show less precision with a deviation of more than 0.55%. The models of Forristall and Garcia-Valladares have been ranked third and last respectively, with absolute uncertainty of 0.782% and 1.202%. The evaluation of the accuracy of the model in predicting the heat losses have been carried out based on 7 test cases of different ambient conditions. At higher operating temperatures, the present model has shown significant improvements. At 350 degrees C, the model has shown an uncertainty similar to that of experimental tests, ie, 7.3W/m(2). Moreover, it has been ranked first with an average deviation of 3.42W/m(2), more accurate than those developed by Behar, Garcia, and Forristall. If the glass of the solar receiver is broken during operation, the present model might offer some knowledge about their performance. Indeed, the model has also been checked for predicting the thermal performance of PTC with air in the annulus of solar receivers. The RMSE is found to be less than 1.5% for the case of efficiency during on-sun tests, and values of 2.292% and 1.562% have been obtained for Garcia-Valladares and Forristall models. Similar finding has been observed for the case of heat losses. Therefore, the proposed model might be an accurate tool for parabolic trough collector performance predictions.