Miller and Richards ( ) were perhaps the first to draw attention to the way in which most reconstruction models related to classical or Romanesque architecture. For the most part, such models focus on stone structures which are often well-understood – churches, temples, bathhouses and the like – and which, with their repeated and regular structural elements, are in some respects relatively easy to construct although the resulting models may be extremely elaborate and complex.
Furthermore, the levels of interpretation required may be relatively limited in some cases, with the monument remains often standing to head height or more. For example, although the reconstruction of the Hadrianic Baths at Leptis Magna was based on a combination of plans and photographs of the ruins, descriptive accounts by modern authors and the evidence of contemporary buildings in other parts of the Roman empire ( ), the structure itself still stands to some height. A similar example is the Temple of Hera II at Poseidonia, a reconstruction of an extremely well-preserved structure, with all columns of the peristyle in situ, and the superstructure preserved up to the horizontal and raking cornices ( ).
This is not to suggest that such reconstructions have no value, simply that a limited degree of interpretation is required in comparison with reconstructions of more 'typical' archaeological remains which are constructed literally from the ground up on the basis of foundation evidence and possibly associated fragmentary building remains. Examples of these include the classic IBM reconstruction of the Old Minster at Winchester ( ) and the reconstruction of Chetro Ketl Great Kiva, New Mexico ( ).
In these and other such cases, the intention is to present a reconstruction of buildings which have only very incomplete remains surviving and which may be difficult for a non-expert audience to understand. Given this, it is noticeable that such reconstructions are still generally of masonry structures of one kind or another, and that the reconstructions of timber and earthwork structures are still extremely rare in comparison. Exceptions include Reilly's fairly limited reconstruction of the Iron Age enclosure at Navan, Northern Ireland, and Ozawa's more elaborate reconstruction of the village of Yoshnigari, Japan (see ).
The reason why such reconstructions are rare is not entirely clear. It may be, for example, that timber structures are perceived as being 'simple' and lacking the elaboration that can be demonstrated to some effect in a computer model. Conversely, it may be that timber structures are perceived as being too complex, in that the irregular nature of their building materials would tend to limit the repeatability of structural elements. In either case, such preconceptions would deny the undoubted complexity of many timber structures and ignore questions about construction and form which apply equally to timber structures and masonry ones.