Oleg (and Jon) OK - if you prefer. As a climber I find value in Jon's somewhat poetic way of adjusting the definition for successfully meeting the criterion that defines the termination of phase3) - namely: "...when you can secure the same response from the mountain your model does - that is, the mountain lets you climb her, via a specific route." I have a beautiful 17 year silver Arab gelding who runs like the wind. The translation of the English sentence "I rode my horse to the watering hole" into Navajo (according to some linguist friends of mine - I haven't the competency at present to investigate directly) is "My horse and I rode to the watering hole." You wrote, "...The issue is that you used books for some kind of preliminary modeling. And the question I would like to get an aswer to is just how do you consider such things?" The difference that makes the difference between NLP modeling (as defined in Whispering) and other forms of modeling is whether the modeler takes all measures possible to remove personal filters (f2 mapping) until they have captured the patterning (criterion). Clearly, any approach that involves f2 mapping prior to achieving criterion will fall short of capturing the patterning of the model selected to the extent it is possible. This is a significant distinction. Note, for example the absence of any models in NLP that have the influence and effectiveness of those that established the field (Bandler and Grinder's work). This is not a consequence of any differences in intelligence or commitment among the modelers involved but rather the difference in method. If you want a model in which your personal filters are mixed up inextrictably with those pattersn offered by the model (the source of the patterning) then go right ahead. I believe that this is the third time I have suggested that you act congruently in both classes of modeling options and THEN offer some comments - I get the feeling I am talking to someone who has no basis in experience for what he is saying. All the best, John