The paper presents THEMIS 2.0, a self-consistent model for dust extinction, emission, and polarisation. The model updates the optical properties of silicates and incorporates new data from laboratory measurements of amorphous silicates, which are expected to be good analogues of diffuse interstellar matter. The model allows for the calculation of polarised extinction and emission, addressing discrepancies in previous models that failed to explain observed variations in extinction and emission. The new optical constants derived from laboratory measurements are used to calculate absorption and scattering efficiencies for spheroidal grains using the discrete dipole approximation. The model accounts for variations in the optical properties of different silicates, which naturally explain the observed variations in extinction and emission. The model is flexible and based on laboratory measurements, making it suitable for understanding the exact nature of interstellar grains and their evolution. The model is also relevant for future cosmic microwave background missions aiming to measure CMB spectral distortions and polarisation. The paper details the new optical constants, the methodology of the optical property calculations, and the variations in optical properties depending on grain shapes and sizes. It also discusses the observational constraints and the definition of the THEMIS 2.0 model. The model is tested against observational data, including the spectral energy distribution and extinction curves, and is found to accurately reproduce observed variations in extinction and emission. The model is also used to study the influence of core-mantle structure on the optical properties of large grains, and it is shown that for the largest grains, the core-mantle structure has a significant impact on the optical properties. The model is further tested against observational constraints on element abundances and dust-to-gas mass ratios, and it is found to be consistent with these constraints. The model is also used to study the influence of grain size and composition on the optical properties of silicate and carbonaceous grains, and it is shown that the optical properties vary significantly with these parameters. The model is found to accurately reproduce the observed variations in extinction and emission, and it is therefore a valuable tool for understanding the properties of interstellar dust.The paper presents THEMIS 2.0, a self-consistent model for dust extinction, emission, and polarisation. The model updates the optical properties of silicates and incorporates new data from laboratory measurements of amorphous silicates, which are expected to be good analogues of diffuse interstellar matter. The model allows for the calculation of polarised extinction and emission, addressing discrepancies in previous models that failed to explain observed variations in extinction and emission. The new optical constants derived from laboratory measurements are used to calculate absorption and scattering efficiencies for spheroidal grains using the discrete dipole approximation. The model accounts for variations in the optical properties of different silicates, which naturally explain the observed variations in extinction and emission. The model is flexible and based on laboratory measurements, making it suitable for understanding the exact nature of interstellar grains and their evolution. The model is also relevant for future cosmic microwave background missions aiming to measure CMB spectral distortions and polarisation. The paper details the new optical constants, the methodology of the optical property calculations, and the variations in optical properties depending on grain shapes and sizes. It also discusses the observational constraints and the definition of the THEMIS 2.0 model. The model is tested against observational data, including the spectral energy distribution and extinction curves, and is found to accurately reproduce observed variations in extinction and emission. The model is also used to study the influence of core-mantle structure on the optical properties of large grains, and it is shown that for the largest grains, the core-mantle structure has a significant impact on the optical properties. The model is further tested against observational constraints on element abundances and dust-to-gas mass ratios, and it is found to be consistent with these constraints. The model is also used to study the influence of grain size and composition on the optical properties of silicate and carbonaceous grains, and it is shown that the optical properties vary significantly with these parameters. The model is found to accurately reproduce the observed variations in extinction and emission, and it is therefore a valuable tool for understanding the properties of interstellar dust.