A quick answer to you question about ligands, Nataraj.
This paper : http://dx.doi.org/10.1016/j.jmb.2003.09.032 Evers, Gohlke, Klebe, Ligand-Supported Homology Modelling of Protein Binding-sites using knowledge based potentials J.Mol.Bio. 334(2) 327-345
Describes the results of blending a ligand-protein docking function with Modeller's objective function. This approach is a shortcut to getting a good conformation for the complex without a full physical treatment of the protein-ligand interaction. The full treatment is possible (can someone correct me here!) if modeller has parameters for both your ligand and its interactions with the protein. Sometimes you can do this by explicitly specifying restraints (hydrogen bonds and salt bridges). Either way, you should always get a better model of the ligand-protein complex because you have used more information in the modelling process.
So your answers : > 1)Wheather my this idea is theoreticaly correct ? Yes.
> 2)if the answer for my question 1 is correct, what is the way or > routine which I can use it for adding my ligand(small > molecule,eg.,DDT) to my model? Look at the FAQ's about parameterising, adding ligands, and setting explicit restraints for things like hydrogen bonding, coordination geometry (metal centers). I know of no single function that will do all this automatically (yet).
hope this helps. j.