Crop models try to simulate the interaction between soil-atmosphere-plant systems with the help of computers. They can be also effective tools in research, education and evaluating the performance of agricultural management practices. This study was aimed to develop a new and efficient crop model based on CSM-CERES-Maize model with ability of considering saline and drought stress that it’s as s pecific condition of arid regions. In this study, a new crop model was tested that integrates its various components with the purpose of developing a more efficient model. In the hybrid model, CERES-Maize supplies SWAP model with daily leaf area index and root depth and on the other hand SWAP model supplies CERES-Maize with daily weather data, soil water content, actual and potential soil evaporation and plant trairation calculated from Penman-Monteith equation and finally the actual root water uptake. For assessing the performance of CERES-Maize-hbased in water deficit conditions, a data set of a field experiment previously carried out on silage corn at the Agricultural Research Center in Varamin, Iran, in 2004 was used. The data set included all required soil, crop, and weather data of the experiment with two deficit levels (0.7 soil moisture deficit (SMD) and 0.85 SMD), full (SMD) and over irrigation treatment and also three nitrogen (N) fertilization levels (0, 150 and 200 kgN/ha). The CERES-Maize-hbased was calibrated with the genetic coefficients that were obtained from previous studies and also the crop coefficients claculated for each irrigation treatment at 150 kgN/ha. Then the calibrated model was evaluated for 0 and 200 kgN/ha treatments for all irrigation levels. Simulation results were compared to the original DSSAT-CERES-Maize model. According to the comparison of the both models in many different aspects, it seems that both models can simulate very well the maize production . A significance difference was found in their prediction of actual soil evaporation, actual plant trairation and average and total simulated biomass but since no significance difference was found in their soil water simulation, this matter became really complicated to decide which one of the models was better . Actually, it needs further studies to check their performance with more details. Despite the fact the new model needs more improvement, generally because the simulation of target parameter (biomass) was fairly successful in both two years, it can be concluded that the model is enough ready for more deep research on simulation of maize growth. Keywords: CSM-CERES-Maize, CERES-Maize-hbased, SWAP, Crop growth models, Maize