I got your message with your coordinates.
Your location actually illustrates the last paragraph in my previous post.
For example, there's the "best" plot at 1' AGL:
http://www.tvfool.com/?option=com_wr...9e74e570576268
And a few more of varying heights:
30 ' AGL, a fairly typical suggestion:
http://www.tvfool.com/?option=com_wr...9e74122219a90a
60', the usual practical height limit for a tower:
http://www.tvfool.com/?option=com_wr...9e742b08e237fc
And the impractical heights of:
200':
http://www.tvfool.com/?option=com_wr...9e740eb0b67e1f
and 500' (limit of TVF):
http://www.tvfool.com/?option=com_wr...9e748a1d8926b3
Which is the most accurate? Probably none of them. You're about 2000' below the ridge line around 3-4 miles to your south-east and, it appears, just at or below the ridge line a half mile to the SE. Any reception that you get will be SEVERELY diffracted and you will find that your reception will vary tremendously based on the weather and ground conditions *up there*.
You can improve your odds of getting channels on real channels 7 & 13 by installing a 10-element high-VHF Yagi aimed midway between those two channels.
For UHF, I doubt you can improve very much more with an off-the shelf UHF antenna. There are several antennas out there with slightly higher UHF gains at the lower half of the band, but the differential is going to be very slight. If you're into experiments, Tom Ballister's site antennahacks.com gives his results of modifications he made to a DB8. I do know he's been working on similar hacks to the 91XG, but I don't recall having seen any results.
You didn't mention what amp you have, but I'd suggest the most sensitive one your budget can afford (lowest noise figure). If it's a single input pre-amp uch as our PA18 or CPA19, combine the two antennas using a Holland UVSJ for the best results. If a separate input pre-amp, set it up for the two antennas.
Bottom line: You're in a particularly challenging location!
Cheers!
PS: KXLY transmits ABC as virtual channel 4.1 and MyN as virtual channel 4.2.