Accurate soil moisture prediction is essential for advancing agricultural productivity, optimizing water resource management, and strengthening climate adaptation strategies, particularly in hydrologically vulnerable regions such as Louisiana. This study investigates the integration of the newly developed OPtical TRApezoid Model (OPTRAM) with conventional moisture-related vegetation indices to enhance soil moisture prediction. A suite of indices, including the Normalized Difference Vegetation Index (NDVI), Normalized Difference Water Index (NDWI), Enhanced Vegetation Index (EVI), Structure Insensitive Pigment Index (SIPI), Atmospherically Resistant Vegetation Index (ARVI), and Normalized Difference Moisture Index (NDMI), was compared against OPTRAM to evaluate predictive capacity. Support Vector Machine with Radial Basis Function (SVM-RBF) kernel and Random Forest (RF) algorithms were employed using Sentinel-2 imagery coupled with Louisiana weather station records. Model performance was validated using statistical metrics (R², RMSE, and MAE). Results revealed that RF achieved MAE = 0.054, RMSE = 0.069, and R² = 0.539, while SVM-RBF achieved MAE = 0.065, RMSE = 0.079, and R² = 0.690. OPTRAM demonstrated superior performance with RF, whereas indices such as NDMI, EVI, SIPI, and NDWI outperformed OPTRAM in moisture prediction with SVM-RBF. These findings highlight the importance of integrating advanced optical models with machine learning approaches to improve soil moisture monitoring and support sustainable land management in Louisiana.