Soil moisture content (SMC) is a crucial component in various environmental studies. Although many models have been proposed for SMC estimation, developing new models for accurate estimation of SMC is still an interesting subject. This study aimed to develop new models for SMC estimation using the water absorption bands in the spectral signatures of three different soil types: loam, silty loam, and sandy loam. Based on the three absorption bands (i.e., 1400, 1900, and 2200 nm) and regression analyses, six approaches were considered. These scenarios were generally based on the reflectance value and its logarithm, as well as the difference between the wet and dry reflectance values for the absorption bands. Finally, 24 models were developed for SMC estimation from the three different soil types, as well as the entire soil samples. The most accurate SMC, as indicated by the lowest root mean squared error (RMSE) and the highest correlation coefficient (r), was obtained from the model developed using the logarithm of the average values reflectance in the three water absorption bands for sandy loam (RMSE = 0.31 g/kg, r = 0.99). Overall, using the spectrometry data derived in the lab, the results of the proposed models were promising and demonstrate great potential for SMC estimation using spectral data collected by satellites in the future studies.