The model potential approach is well adapted to study atomic and molecular systems involving a single active electron. Such is the case of the alkali-dimer lithium cation Li + 2. However, a comparison of the model potential results of Magnier et al.[1] and those based on ab-initio techniques [2, 3, 4] raises a number of questions related to the existence of an important disagreement regarding several excited states,which are found to be repulsive by Magnier et al. [1] but attractive when ab-initio techniques are employed. In this paper, we propose to re-investigate the Li + 2 system, using a model potential technique to compute the adiabatic energy curves and the molecular spectroscopic constants. Our aim is to clarify whether this disagreement between the ab-initio and model potential methods originates from some conceptual defect of the model potential technique or whether there is some source of error in the calculations of Magnier et al. [1].