I use AMR for healing sims, though I don’t play disc actively (resto druid mainly). AMR has moved away from the idea of stat weights generally, as they can be misleading or only locally valid, but I get very similar results using the Pawn export function on my old priest when optimising for Disc.
If i remember correctly, the Disc community used to value haste very strongly, basing that on some pretty fuzzy math and assumptions. In the spreadsheets shared back when I was playing my priest alt, the Haste “stat weight” was determined by taking the highest of all of your other weights, and multiplying it by 1.1, to make sure that haste was always the most valued stat. I’m not sure how the priest community views it now, but i wouldn’t be surprised based on the common knowledge back then if Haste wasn’t really quite valuable. Whether 20/9/9/8/8 is “right” even locally I can’t really say for sure, but I’d say the general order and magnitude doens’t seem unreasonable at a glance.
Additionally, there might be some other factors that could elevate haste’s value in AMR’s current strategy specifically. The fight that the Raid strategies are currently based on is Krosus from Nighthold, which is a combination of heavy tank damage and really bursty raid damage. For really bursty raid damage, it seems reasonable that haste is particularly valuable for Disc, as getting in one or 2 more smites or an extra penance before the raid is topped for each burst would be a pretty massive gain, as would being able to get atonement on one or two more folks before the Slam/Bomb went out.
Long story short, it doesn’t seem off to me at a glance, but you can always check by simming both your current gear (or your set optimised based on what you think might be best) and the recommended changes from AMR and see what happens. If you would like to share your export from the AMR addon I can take a look as well.