diff --git a/TODO.md b/TODO.md
index 01846a8..0dcc836 100644
--- a/TODO.md
+++ b/TODO.md
@@ -1,5 +1,5 @@
 - (!) README.md
-- Do not use OrderedDict (dictionaries in Python 3.7 are now ordered)
+- Docstrings
 - Features
     - Models layer
         - [X] Data variant converting
@@ -7,6 +7,7 @@
         - [X] Data variant merging
         - [X] Create and register models
         - [X] Merge meta-models and extract common ones
+            - [ ] Strict mode
         - [X] Save meta-models as python code
             - [X] typing code generation
                 - [ ] (Maybe in future) Extract to another module (by serializers for each dynamic typing class)
diff --git a/json_to_models/dynamic_typing/typing.py b/json_to_models/dynamic_typing/typing.py
index cd93c22..e6e0ba6 100644
--- a/json_to_models/dynamic_typing/typing.py
+++ b/json_to_models/dynamic_typing/typing.py
@@ -1,5 +1,4 @@
 import operator
-from collections import OrderedDict
 from inspect import isclass
 from typing import Dict, Set, Tuple
 
@@ -28,7 +27,7 @@ def compile_imports(imports: ImportPathList) -> str:
     """
     Merge list of imports path and convert them into list code (string)
     """
-    class_imports_map: Dict[str, Set[str]] = OrderedDict()
+    class_imports_map: Dict[str, Set[str]] = {}
     package_imports_set: Set[str] = set()
     for module, classes in filter(None, imports):
         if classes is None:
@@ -42,7 +41,7 @@ def compile_imports(imports: ImportPathList) -> str:
             class_imports_map[module] = classes_set
 
     # Sort imports by package name and sort class names of each import
-    class_imports_map = OrderedDict(sorted(
+    class_imports_map = dict(sorted(
         ((module, sorted(classes)) for module, classes in class_imports_map.items()),
         key=operator.itemgetter(0)
     ))
diff --git a/json_to_models/generator.py b/json_to_models/generator.py
index fa563f0..239a0a7 100644
--- a/json_to_models/generator.py
+++ b/json_to_models/generator.py
@@ -1,5 +1,4 @@
 import re
-from collections import OrderedDict
 from typing import Any, Callable, List, Optional, Pattern, Union
 
 from unidecode import unidecode
@@ -114,7 +113,7 @@ def merge_field_sets(self, field_sets: List[MetaData]) -> MetaData:
         """
         Merge fields sets into one set of pairs (key, metadata)
         """
-        fields: dict = OrderedDict()
+        fields: dict = {}
 
         first = True
         for model in field_sets:
@@ -166,7 +165,7 @@ def optimize_type(self, meta: MetaData, process_model_ptr=False) -> MetaData:
             Default is False to prevent recursion cycles.
         """
         if isinstance(meta, dict):
-            fields = OrderedDict()
+            fields = {}
 
             for k, v in meta.items():
                 fields[k] = self.optimize_type(v)
diff --git a/json_to_models/registry.py b/json_to_models/registry.py
index 4818392..c7c37de 100644
--- a/json_to_models/registry.py
+++ b/json_to_models/registry.py
@@ -1,4 +1,4 @@
-from collections import OrderedDict, defaultdict
+from collections import defaultdict
 from itertools import chain, combinations
 from typing import Dict, Iterable, List, Set, Tuple
 
@@ -50,7 +50,7 @@ def __init__(self, *models_cmp: ModelCmp):
         :param models_cmp: list of model comparators. If you want merge only equals models pass ModelFieldsEquals()
         """
         self._models_cmp = models_cmp or self.DEFAULT_MODELS_CMP
-        self._registry: Dict[str, ModelMeta] = OrderedDict()
+        self._registry: Dict[str, ModelMeta] = {}
         self._index = Index()
 
     @property