Analytical models are very useful to the vehicle dynamics project engineer in order to provide simple solutions that bring at the same time a deep understanding of the physical phenomena being studied. Due to their structure, the input of analytical models are only the variables of interest affecting the concerned handling metrics - this fact reduce the parameters quantity to build the same model, making them proper choices for advanced studies. Additionally, analytical models are more efficient in computational terms, aspect convenient for studies that involve large amounts of calculation iterations like numerical optimization processes. This paper presents analytical model results for the roll gradient, understeer gradient and steering sensitivity metrics and proposes enhancements in order to obtain results with accuracy compatible to the experimental measured variables. Finally, the performance of these enhanced analytical models is compared against experimental measurements and more detailed multibody models.