The answer to all the three points is randomness and symmetry.
So what the molecule represents in fig 24.1 in your derivation is nothing but the representation of the average motion of all the particles in your system. This is based on randomness. Think of it this way - if the average molecule did not perform the motion as given in the derivation, what else could it have performed?
why collisions with other molecules not taken into account since they could also influence the time it takes for one molecule to collide with the wall
They would influence the time. However, if one molecule's time to collide is increased by $"t"$, you would expect another molecule's to be decreased by $"t"$. If you average out the time over the number of molecules, you would get no net deviation in time.
how can we assume that the velocity of particle in x y z direction is same i mean how can this be called a good approximation?
Again the molecule in the derivation represents average motion. So if you are talking about averages, what special direction can you select where speed should be different? Note that all X, Y and Z are equivalent and there is no reason for us to treat anyone preferentially.
how can taking average force into account and not the real forces the equation still works very well
Again the average force represents the mean of "real" forces.
Note that this is not a strict derivation but only an attempt to justify the known results.