Common biology ID codes.
PDB: 2BIB for example. 4 letter code for a specific structure.
Uniprot: Q9NZT1 for example. A six letter combination of numbers and letters for either a TrEMBL or Swissprot record. This record contains information on domains, and Uniprot links out to lots of useful things like interaction databases, redundancy datasets etc.
Pfam: X1WG39_DANRE A uniprot ID followed by an identifier. Tightly linked with Uniprot, but with an emphasis on domains.
NCBI: NP_000108.1 For example. These IDs link to either a gene, transcript or protein in refseq. In an NCBI record there are two main distinctions in prefix of the ID code: hypothetical/automatic assertions (XM_, XR_, and XP_) and manually curated assertions (NM_, NR_, and NP_). This is followed by a 6 digit number, and sometimes followed by a "." and a number, indicating an splice isoform.
I'm not sure where you got this output so this seems a strange concatenation of Uniprot information.
Q9NZT1 is indeed the unique Uniprot ID. This is specific to uniprot.
CALL5_HUMAN is the "gene name" or "Identifier" and often spans across different databases for that gene. It's a little more back end, but is essentially not unique due to splice isoforms.
There are many IDs in uniprot, and many types. Already below you'll notice some are swiss prot, others are Trembl. These are the main types. Swissprot are manually reviewed, whereas Trembl is an automatically compiled collection.
There are many many different accession code types with each database using it's own nomenclature that you'll come across. Unfortunately, there are too many to list here comprehensively.