The current existing researches were discussed to assess the advantages and limitations of existing empirical and theoretical models for direct conductivity (DC) prediction. Prior research has demonstrated that an effective medium model may not accurately reflect the actual situation due to the presence of two types of water (bound water and bulk water) in clay-rich materials. Furthermore, the existing models can not satisfy the prediction of electrical conductivity of metal ions adsorbed clay or unsaturated clay. To address this issue, a new Effective Medium Double Water (EMDW) model was proposed based on multiple scattering techniques, which encompasses soil particles, surface-bound water layers, and bulk water and was established by controlled soil types and degrees of saturation. The novel EMDW model includes the Coherent Potential Approximation (CPA), which has consistently demonstrated superior agreement with experimental data when compared to other approximation models. Moreover, the binomial expansion approximation was utilized to simplify the formula and facilitate its use. The developed conductivity model was validated with data from other researchers. In comparison to other well-established conductivity models, the proposed EMDW model has clear physical meaning and can accurately compute matrix conductivity utilizing modified coated particle conductivity and saturation conductivity. The findings suggest that matrix conductivity in clay materials is significantly correlated with electrical-physical parameters, such as porosity, degree of saturation, shape of each discontinuous phase, and conductivity of surface-bound water and bulk water. Consequently, the new EMDW model is a theoretically grounded, physically meaningful, and easy-to-use model for conductivity prediction in clay materials.