The maths in the linked articles is annoyingly wrong.
For example in the stackoverflow-answer it says
Required Acceleration = LOS * LOS_Rate * NC + APN_bias and
APN_bias = LOS_Rate/delta_T * ( NC/2 ). So since LOS is a vector and all the other terms are scalars that would mean adding a vector and a scalar.
And in the article on moddb it says
# Now, calculate the final lateral acceleration required for our missile
# to home into our target.
latax = RTM_new * N * Vc * LOS_Rate + LOS_Delta * Nt * ( 0.5 * N )
So the formula is supposed to result in a LATERAL acceleration - but the second part of the addition contains the factor Nt which is supposed to describe the relative acceleartion of the target - so if the target moves without acceleration it's 0. (The article talks about flying missiles and therefore factors in gravity, but let's talk about the special case of movement in a 2D plane first to check the formula - after all it should work just as well in that special case.
So this leaves us with
latax = RTM_new * N * Vc * LOS_Rate
and again RTMnew is a vector and all the other terms are scalars. So the result will have the same direction as RTMnew, and that vector is pointing directly at the target, so we end up accelerating at the target, not at a point in front of the target.
I'll try to find the time and test it myself, but these two sources seem too faulty to simply copy and modify.