The NIST PQC project had started since 2016, deadline for entry had passed on November 2017. But according to NIST-IR-8309, they are still open to new ideas. So what are they looking for?
They said this (on page 26):
NIST is pleased with the progress of the PQC standardization effort but recognizes that current and future research may lead to promising schemes which were not part of the NIST PQC Standardization Project. NIST may adopt a mechanism to accept such proposals at a later date. ...
We could only guess what they may be planning.
First, the 3 Lattice-based KEMs, (Kyber, NTRU, Saber) are as efficient as physically and logically possible - the maths they use spends little time compared to the hash and XOF functions they use as components. So if you have an idea on a key establishment algorithm, it'd better be significantly more efficient and more secure than those 3, and should ideally be based on something other than lattice.
Also, their codes are public. You can compile and benchmark them to see if your algorithm runs any faster than any of those 3.
Secondly,
... In particular, NIST would be interested in a general-purpose digital signature scheme which is not based on structured lattices.
Because the only digital signature schemes compact enough for general use are Dilithium and Falcon. Dilithium is based on modular lattice (somewhatly structured lattice) and Falcon is based on NTRU lattice (very structured lattice).
So if you have a miracle signature algorithm that's as efficient as those two, and have compact public key and signature size and is based on somewhatly different hardness assumption, NIST would have every reason to consider it in the future.