First, I should apologize both to you and the others who have tried to help you.  If I'm going to throw potshots at other's comments I should at least address the original question, too. 

I'd have to know about your area in order to anwer your questions very well.  

Here are some things to think about:

What is the source of varibility in your fields?  How much varibility is there?  What is your philosophy as you approach this, are you looking at it as a remedial action or as a loss stopping action?  What are you're fertilizer costs?  Are you dealing with long-term land or year-to-year?

Based on my experience, on my farms, I believe VR lime is a worthwhile practice.  Our soil types are very varible.  We often have high varibility in pH.  We have high variblibly in CEC.  These combine to create EXTREMELY high variblility in lime recommendations.   

P and K, on the other hand, I find to be more problematic when it comes to justifying VR.  In my analysis it seems to be more profitible on my farms to simply invest any money which would be considered for VR work in higher rates of P and K.  I really do not think, in my experience, that P and K are very good places to invest in VR.   I believe that N and population have much more potential from VR then P or K do. 

First mistake which i think many make is thinking they are going to "fix" things.  I find that most of our low yield situations are not fixable.  Many initially start with the idea of "getting the low yield areas up" but in fact these areas are already reaching their potential, but the high yielding areas quite likely have needs which are not being met, which could cause them to be even higher yielding in the future.  Another potential is to reduce costs on those low areas. 

I firmly believe that you need to understand the processes you apply.   If you have been in the lab you have a different expectation from soil testing then many have.  Contrary to what you might like to believe, it is not an infallible process.  There is alot of opportunity to get different results even from an identical sample.  I feel that many treat soil test results as abolute values but I do not think that is realistic.  The same thing applies to the equipment to be used, both mechanically as well as electronically.  I've seen countless hours wasted in discussions about "boundary areas" which are prefaced on the ideas that 1)we have an accurate idea of what we are dealing with in the field and 2)we are using equipment which acts in an almost binary manner.  Reality says neither is true.  We don't know that point 'a's value is exactly 'x' and when we tell our equipment to apply 'y' product at point 'c' what actually happens is that it heads toward that rate sometime in an attempt to be there at the point, but it's not a binary process where at point 'b' it is this and at point 'c' it is instantly at the next level.  Knowing how the equipment works makes the futility of the discussion apparent.

Honestly, if you pay someone else to do everything the situations where it is economic are fairly uncommon.  In order to make it pay, in my opinion based on my experience and location, you have to do it on a scale which allows costs to be contained and also to be spread over sufficient acres.   

I'm sorry if this is a ramblefest and if there are spelling or syntax errors.  I shouldn't write at this hour.