I don't know where the hell you got these inflated numbers for dead Red Army personell. The number of Soviet troops that engaged the Axis in that battle was 1.2 million. They won the battle. They didn't loose every third man, let alone suffer casualties at 85% of their force. Operation Citadel was postponed when troops were needed in Italy. If Germans don't commit to attacking the Soviets at Kursk and biting into their trap and instead keep the forces in Italy, Stalin would have those extra troops to spare for an Italian adventure.
Red Army forces at Kursk, including both Operation Citadel (Red Army forces deployed 1,910,361 with KIA/MIA 70,240 i.e. unrecoverable losses, 104.714 WIA and the Soviet counter-offensives Operations Kutuzov with 1,286,000 troops deployed (112,529 KIA/MIA i.e. unrecoverable losses, 317,361 WIA) & Polkovodets Rumyantsev with 1,114000 troops deployed (71,611 KIA/MIA i.e. unrecoverable losses, 113,995 WIA) totaled ~2.5 million men (there was considerable replacement activity that makes determining the exact figure engaged almost impossible). Total losses, even those detailed to the man by Soviet records, are almost certainly understated/cooked (as an example the number of unrecoverable losses suffered by the Steppe Front and the Voronezh Front during Citadel are identical and end with a round figure) but do total, even in the official records, as 254,529 KIA/MIA ( i.e. unrecoverable losses), 536,070 WIA or a total of 790,599 ground force losses. This does not include losses by Soviet air formations. It is worth noting that these losses are also unlikely to include even a majority of the "penal battalion" losses.
In short the Soviets did, very much suffer casualties of at least one third of the deployed force, with official KIA numbers almost certainly suppressed and "died of Wounds" not included in the KIA figures.
Sources:
Krivosheev, Grigoriy
Soviet Casualties and Combat Losses in the Twentieth Century
Beevor, Antony
, The Second World War