A simple model that can assess the river rehabilitation using rehabilitation potential index (RPI) of fish habitat has been developed. This model calculates the RPI using three indicators: the habitat suitability index (HSI), niche breadth (NB), and niche overlap (NO). These indicators were estimated by using environmental factors that affect fish habitat in the river and then RPI was obtained by combining the indicators. However, the model showed some limitations. For example, the representative NO was obtained by summing each NO, which does not reflect the difference of the number of fish species in rivers. This study improved the estimation method of NO by using the averaged NO that can consider different kinds of species in each river. In addition, the previous study used standardization method for making three indicator values in the range of 0 to 1. However, the standardized indicators can distort the relative size when we compare with the original data. In this study, we have selected alternative methods to calculate indicators, eliminating the need for standardization. Along with these improvements, this study developed a new RPI estimation model by reviewing various methods and selecting the proper ones. The RPIs of the two models were estimated using fish abundance and habitat environmental factors from 57 sites in Han river, Geum river, and Nakdong river in Korea. We compared the results of the existing model with the new model using root mean square error (RMSE) and mean absolute deviation (MAD). The new model showed the reduced uncertainty and enhanced applicability in comparison to the previous model.