Previous works on the visual servoing of parallel robots using the observation of their leg directions validated the feasibility of the approach but they have enlighten two main surprising results for which no answer was given: (i) the observed robot which is composed of n legs could be controlled in most cases using the observation of only m leg directions (m < n), and that (ii) in some cases, the robot did not converge to the desired end-effector pose, even if the observed leg directions did (i.e. there was not a global diffeomorphism between the observation space and the robot space). Recently, it was shown that the visual servoing of the leg directions of the Gough-Stewart platform and the Adept Quattro with 3 translational degrees of freedom was equivalent to controlling other virtual robots that have assembly modes and singular configurations different from those of the real ones. These hidden robot models are tangible visualizations of the mapping between the observation space and the real robots Cartesian space. Thanks to this concept, all the aforementioned points were answered for the mentioned robots. In this paper, the concept of hidden robot model is generalized for any type of parallel robots controlled using visual servos based on the observation of the leg directions. It is shown that the concept of hidden robot model is a powerful tool that gives useful insights about the visual servoing of robots using leg direction observation. With the concept of hidden robot model, the singularity problem of the controller can be addressed and the convergence issues of the controller can be explained, understood and solved. All these results are validated in simulations and through experiments on a Quattro robot.