He argues that while these transactions break the common-input-ownership heuristic, they are not fundamentally different from standard CoinSwap transactions in terms of coin ownership analysis. However, he does acknowledge that PayJoin-with-CoinSwap could be useful for individuals trying to recover degraded privacy, particularly if they are spending from a reused or otherwise linked address. The mere existence of PayJoin-with-CoinSwap could also improve privacy for those who don't use it, as it would make it more difficult for an adversary to assume ownership of co-spent coins.Belcher also notes some downsides to PayJoin-with-CoinSwap, including the possibility of spying and the need for makers to reveal additional UTXOs during the swap protocol. He suggests using decoy UTXOs rather than PoDLE to resist attacks, as this would allow the probability of an attacker guessing the maker's real UTXOs to be lowered. Additionally, Belcher argues that funding transactions for chained/routed swaps could all be in the same block, contrary to what some have claimed.Finally, Belcher suggests exploring whether private key handover could work with S6, another type of CoinSwap transaction. This would allow for off-chain transfer of coin possession and potentially improve privacy.