The prediction of shear resistance in reinforced concrete (RC) beams, which are strengthened using externally bonded reinforcements (EBR), poses a significant challenge due to the complex nature of shear-resisting mechanisms and their complex interaction. Conventional models have been shown to have limited reliability and produce weak predictions. However, recent advancements in machine learning techniques and artificial intelligence have led to the development of several models in various studies that aim to predict the contribution of EBR-FRP to the shear resistance of RC beams. This study makes an attempt to enhance the predictive performance of the models by considering various factors. First, a comprehensive review is conducted on previous ML models, with a discussion on their strengths and weaknesses. The limitations of these models are addressed. Furthermore, potential approaches to improve the model predictive performance are discussed, and a new model is proposed.