Choir live recording: mic placement recommendations?

Hello fellow Sonarites.
 
I will be recording a big "classical" choir (with organ) in a church in a couple of weeks.
 
I am planning to use 3 condenser microphones to do the job.(One At-2050 and two AT-3031).
 
But, as I am more of a "close micking" / rock - popular music kind of tech, I would like to have some advice from fellow producers.
 
So, I am thinking of placing the AT-2050 about 20 feet in front of the choir (dead center).
 
Afterwards, I am thinking of placing the AT-3031 microphone pair (XY position, dead center, about 6-8 feet behind the first mic...)
 
OR
 
Place 2 of my mics to record using MS techniques.
 
EDIT: I'm realizing I should be more specific here. In this context, I am dealing with an artificial "organ" that goes through the PA system. Now, in order to attempt to respect as much as possible the balance between the choir and the organ that the conductor is aiming for, I would like NOT to record the organ on an independant track. That is why my approach here is considering mic positions that are not THAT close to the choir, in order to capture the organ's sound going out of the P.A .
 
SO that's why i'm a considering mic placements at 2 different distance from the performers... I'll try to match as best as I can the balance I'm going to hear live by balancing my 3 tracks. What phase-cancelation pitfalls should I be carefull about here?
 
For the stereo pair, should I consider and ORTF technique instead of the XY one in that context?
 
Recommendations appreciated. Thanks!

"You may need some good height from the boom" Read "will need some good height from the boom."

Thank you all for your replies.
 
I never thought an XY recording pattern is not used to record ensembles... I'll look it up but I am surprised. ORTF... used it in the past (have a stereo "bar" to put my two mics on...)  I'll have to be careful about the angle.
 
The third mic is more aimed to be a "closer mic" than a "soloist" mic   ... But I'm considering this approach only because of the organ... ("classical" musicians... at least the ones I know... are very demanding about their "balance" being as natural as can be... hence the idea NOT to record the organ on a separate track... well, I guess I'll record it anyway, i'll just try not to use it...)
 
Found out about a technique named the DECCA TREE, which is looking pretty much to be the "closest fit" to my original approach as can be... only difference in my case being... Only one of my tree mics can be omni-directional.... So I guess I could try it anyway, using only cardoids...
 
I'll now have a look at the links posted by Rimshot. Thanks again.

Found out about a technique named the DECCA TREE, which is looking pretty much to be the "closest fit" to my original approach as can be... only difference in my case being... Only one of my tree mics can be omni-directional.... So I guess I could try it anyway, using only cardoids...

 
Decca is a troublesome approach.  It was designed for omnis, but more specifically the Neumann M50 which is a small diaphragm capsule mounted in a ~50mm ball.  Usually to replicate this you need a modern microphone of the same design like the TLM50 or M150, or DPA, Schoeps, Gefell and Josephson microphones which include a spherical baffle as an accessory.  Another problem, is without outriggers (additional omni microphones 7-8feet to the left and right) the stereo image is unstable at different volumes.  
 
The problem with XY is the SRA (recording angle) is too wide, which equates to a very narrow playback image.  It is also purely coincident so there is no time delay between mics which is needed for a realistic sense of space.  The pattern works better on large groups with hypercardioid microphones, however.