Use the two-byte form for larger values

Read literal lengths and independent pad bytes; ID 256 is not an eight-bit element ID.

• The two-byte form uses outer values 0x1000–0x100F: twelve fixed bits 0x100 followed by four application-dependent appbits. An element has an 8-bit ID 1–255 and an 8-bit data length 0–255. The length excludes both header bytes; unlike the one-byte form, do not add one (§4.3).
• The chosen zero-length element ID 200 is c 8 00, followed here by two block-padding zero bytes. Its length byte 00 is part of the element header; the later zero bytes are independent padding. The two-byte form can use ID 15 as an ordinary ID;15 is reserved only in the one-byte form (§§4.1.2,4.3).
• Each pad byte is ignored on its own, without supplying an ID or a length. Thus body 00 11 01 ab is one leading pad byte followed by ID 17, literal length 1 and dataab. The 11 is not a length belonging to the pad byte (§4.1.2).
• Signaling ID 256 refers to the four appbits, not an element serialized as ID 0. If no definition for 256 is specified through setup/configuration, the sender SHOULD set appbits to zero and the receiver MUST ignore them. Appbits 0xA without such a definition does not create a new mapped element or require discarding all other elements (§4.3).
• A receiver without an application handler for a declared element can use its structural length to skip that opaque data in this teaching model, then read a later handled element. That is a chosen parsing policy, not a new security guarantee. The helper checks the declared block boundary and never borrows media-payload bytes to complete a truncated element. Local-ID meanings still require out-of-band definition; URI/offer exceptions are explained next.